Showing 1–50 of 151 results for author: Stärk, H

Mechanical work extraction from an error-prone active dynamic Szilard engine

Authors: Luca Cocconi, Paolo Malgaretti, Holger Stark

Abstract: Isothermal information engines operate by extracting net work from a single heat bath through measurement and feedback control. In this work, we analyze a realistic active Szilard engine operating on a single active particle by means of steric interaction with an externally controlled mechanical element. In particular, we provide a comprehensive study of how finite measurement accuracy affects the… ▽ More Isothermal information engines operate by extracting net work from a single heat bath through measurement and feedback control. In this work, we analyze a realistic active Szilard engine operating on a single active particle by means of steric interaction with an externally controlled mechanical element. In particular, we provide a comprehensive study of how finite measurement accuracy affects the engine's work and power output, as well as the cost of operation. Having established the existence of non-trivial optima for work and power output, we study the dependence of their loci on the measurement error parameters and identify conditions for their positivity under one-shot and cyclic engine operation. By computing a suitably defined information efficiency, we also demonstrate that this engine design allows for the violation of Landauer's bound on the efficiency of information-to-work conversion. Notably, the information efficiency for one-shot operation exhibits a discontinuous transition and a non-monotonic dependence on the measurement precision. Finally, we show that cyclic operation improves information efficiency by harvesting residual mutual information between successive measurements. △ Less

Submitted 14 May, 2025; originally announced May 2025.

Comments: 13 pages, 6 figures

Protein FID: Improved Evaluation of Protein Structure Generative Models

Authors: Felix Faltings, Hannes Stark, Tommi Jaakkola, Regina Barzilay

Abstract: Protein structure generative models have seen a recent surge of interest, but meaningfully evaluating them computationally is an active area of research. While current metrics have driven useful progress, they do not capture how well models sample the design space represented by the training data. We argue for a protein Frechet Inception Distance (FID) metric to supplement current evaluations with… ▽ More Protein structure generative models have seen a recent surge of interest, but meaningfully evaluating them computationally is an active area of research. While current metrics have driven useful progress, they do not capture how well models sample the design space represented by the training data. We argue for a protein Frechet Inception Distance (FID) metric to supplement current evaluations with a measure of distributional similarity in a semantically meaningful latent space. Our FID behaves desirably under protein structure perturbations and correctly recapitulates similarities between protein samples: it correlates with optimal transport distances and recovers FoldSeek clusters and the CATH hierarchy. Evaluating current protein structure generative models with FID shows that they fall short of modeling the distribution of PDB proteins. △ Less

Submitted 12 May, 2025; originally announced May 2025.

Future Circular Collider Feasibility Study Report: Volume 2, Accelerators, Technical Infrastructure and Safety

Authors: M. Benedikt, F. Zimmermann, B. Auchmann, W. Bartmann, J. P. Burnet, C. Carli, A. Chancé, P. Craievich, M. Giovannozzi, C. Grojean, J. Gutleber, K. Hanke, A. Henriques, P. Janot, C. Lourenço, M. Mangano, T. Otto, J. Poole, S. Rajagopalan, T. Raubenheimer, E. Todesco, L. Ulrici, T. Watson, G. Wilkinson, A. Abada , et al. (1439 additional authors not shown)

Abstract: In response to the 2020 Update of the European Strategy for Particle Physics, the Future Circular Collider (FCC) Feasibility Study was launched as an international collaboration hosted by CERN. This report describes the FCC integrated programme, which consists of two stages: an electron-positron collider (FCC-ee) in the first phase, serving as a high-luminosity Higgs, top, and electroweak factory;… ▽ More In response to the 2020 Update of the European Strategy for Particle Physics, the Future Circular Collider (FCC) Feasibility Study was launched as an international collaboration hosted by CERN. This report describes the FCC integrated programme, which consists of two stages: an electron-positron collider (FCC-ee) in the first phase, serving as a high-luminosity Higgs, top, and electroweak factory; followed by a proton-proton collider (FCC-hh) at the energy frontier in the second phase. FCC-ee is designed to operate at four key centre-of-mass energies: the Z pole, the WW production threshold, the ZH production peak, and the top/anti-top production threshold - delivering the highest possible luminosities to four experiments. Over 15 years of operation, FCC-ee will produce more than 6 trillion Z bosons, 200 million WW pairs, nearly 3 million Higgs bosons, and 2 million top anti-top pairs. Precise energy calibration at the Z pole and WW threshold will be achieved through frequent resonant depolarisation of pilot bunches. The sequence of operation modes remains flexible. FCC-hh will operate at a centre-of-mass energy of approximately 85 TeV - nearly an order of magnitude higher than the LHC - and is designed to deliver 5 to 10 times the integrated luminosity of the HL-LHC. Its mass reach for direct discovery extends to several tens of TeV. In addition to proton-proton collisions, FCC-hh is capable of supporting ion-ion, ion-proton, and lepton-hadron collision modes. This second volume of the Feasibility Study Report presents the complete design of the FCC-ee collider, its operation and staging strategy, the full-energy booster and injector complex, required accelerator technologies, safety concepts, and technical infrastructure. It also includes the design of the FCC-hh hadron collider, development of high-field magnets, hadron injector options, and key technical systems for FCC-hh. △ Less

Submitted 25 April, 2025; originally announced May 2025.

Comments: 627 pages. Please address any comment or request to [email protected]

Report number: CERN-FCC-ACC-2025-0004

Future Circular Collider Feasibility Study Report: Volume 3, Civil Engineering, Implementation and Sustainability

Authors: M. Benedikt, F. Zimmermann, B. Auchmann, W. Bartmann, J. P. Burnet, C. Carli, A. Chancé, P. Craievich, M. Giovannozzi, C. Grojean, J. Gutleber, K. Hanke, A. Henriques, P. Janot, C. Lourenço, M. Mangano, T. Otto, J. Poole, S. Rajagopalan, T. Raubenheimer, E. Todesco, L. Ulrici, T. Watson, G. Wilkinson, P. Azzi , et al. (1439 additional authors not shown)

Abstract: Volume 3 of the FCC Feasibility Report presents studies related to civil engineering, the development of a project implementation scenario, and environmental and sustainability aspects. The report details the iterative improvements made to the civil engineering concepts since 2018, taking into account subsurface conditions, accelerator and experiment requirements, and territorial considerations. I… ▽ More Volume 3 of the FCC Feasibility Report presents studies related to civil engineering, the development of a project implementation scenario, and environmental and sustainability aspects. The report details the iterative improvements made to the civil engineering concepts since 2018, taking into account subsurface conditions, accelerator and experiment requirements, and territorial considerations. It outlines a technically feasible and economically viable civil engineering configuration that serves as the baseline for detailed subsurface investigations, construction design, cost estimation, and project implementation planning. Additionally, the report highlights ongoing subsurface investigations in key areas to support the development of an improved 3D subsurface model of the region. The report describes development of the project scenario based on the 'avoid-reduce-compensate' iterative optimisation approach. The reference scenario balances optimal physics performance with territorial compatibility, implementation risks, and costs. Environmental field investigations covering almost 600 hectares of terrain - including numerous urban, economic, social, and technical aspects - confirmed the project's technical feasibility and contributed to the preparation of essential input documents for the formal project authorisation phase. The summary also highlights the initiation of public dialogue as part of the authorisation process. The results of a comprehensive socio-economic impact assessment, which included significant environmental effects, are presented. Even under the most conservative and stringent conditions, a positive benefit-cost ratio for the FCC-ee is obtained. Finally, the report provides a concise summary of the studies conducted to document the current state of the environment. △ Less

Submitted 25 April, 2025; originally announced May 2025.

Comments: 357 pages. Please address any comment or request to [email protected]

Report number: CERN-FCC-ACC-2025-0003

Future Circular Collider Feasibility Study Report: Volume 1, Physics, Experiments, Detectors

Authors: M. Benedikt, F. Zimmermann, B. Auchmann, W. Bartmann, J. P. Burnet, C. Carli, A. Chancé, P. Craievich, M. Giovannozzi, C. Grojean, J. Gutleber, K. Hanke, A. Henriques, P. Janot, C. Lourenço, M. Mangano, T. Otto, J. Poole, S. Rajagopalan, T. Raubenheimer, E. Todesco, L. Ulrici, T. Watson, G. Wilkinson, P. Azzi , et al. (1439 additional authors not shown)

Abstract: Volume 1 of the FCC Feasibility Report presents an overview of the physics case, experimental programme, and detector concepts for the Future Circular Collider (FCC). This volume outlines how FCC would address some of the most profound open questions in particle physics, from precision studies of the Higgs and EW bosons and of the top quark, to the exploration of physics beyond the Standard Model.… ▽ More Volume 1 of the FCC Feasibility Report presents an overview of the physics case, experimental programme, and detector concepts for the Future Circular Collider (FCC). This volume outlines how FCC would address some of the most profound open questions in particle physics, from precision studies of the Higgs and EW bosons and of the top quark, to the exploration of physics beyond the Standard Model. The report reviews the experimental opportunities offered by the staged implementation of FCC, beginning with an electron-positron collider (FCC-ee), operating at several centre-of-mass energies, followed by a hadron collider (FCC-hh). Benchmark examples are given of the expected physics performance, in terms of precision and sensitivity to new phenomena, of each collider stage. Detector requirements and conceptual designs for FCC-ee experiments are discussed, as are the specific demands that the physics programme imposes on the accelerator in the domains of the calibration of the collision energy, and the interface region between the accelerator and the detector. The report also highlights advances in detector, software and computing technologies, as well as the theoretical tools /reconstruction techniques that will enable the precision measurements and discovery potential of the FCC experimental programme. This volume reflects the outcome of a global collaborative effort involving hundreds of scientists and institutions, aided by a dedicated community-building coordination, and provides a targeted assessment of the scientific opportunities and experimental foundations of the FCC programme. △ Less

Submitted 25 April, 2025; originally announced May 2025.

Comments: 290 pages. Please address any comment or request to [email protected]

Report number: CERN-FCC-PHYS-2025-0002

E. coli bacterium tumbling in bulk and close to surfaces: A simulation study

Authors: Pierre Martin, Tapan Chandra Adhyapak, Holger Stark

Abstract: Motility is fundamental to the survival and proliferation of microorganisms. The E. coli bacterium propels itself using a bundle of rotating helical flagella. If one flagellum reverses its rotational direction, it leaves the bundle, performs a polymorphic transformation, and the bacterium tumbles. The E. coli bacterium is hydrodynamically attracted to surfaces. This prolongs its residence time, wh… ▽ More Motility is fundamental to the survival and proliferation of microorganisms. The E. coli bacterium propels itself using a bundle of rotating helical flagella. If one flagellum reverses its rotational direction, it leaves the bundle, performs a polymorphic transformation, and the bacterium tumbles. The E. coli bacterium is hydrodynamically attracted to surfaces. This prolongs its residence time, while tumbling facilitates surface detachment. We develop a model of E. coli that uses an extended Kirchhoff rod theory to implement flagellar flexibility as well as different polymorphic conformations and perform hydrodynamic simulations with the method of multiparticle collision dynamics (MPCD). To establish a reference case, we determine the distribution of tumble angles in the bulk fluid, which shows good agreement with experiments for a fixed tumble time. Increasing the hook stiffness, narrows the tumble angle distribution and reduces the flagellar dispersion during tumbling. Close to a bounding surface, the tumble angle distribution is shifted to smaller angles, while flagellar dispersion is reduced. Reorientation within the plane favors the forward direction, which might be an explanation for prolonged run times observed in experiments △ Less

Submitted 29 April, 2025; originally announced April 2025.

ProtComposer: Compositional Protein Structure Generation with 3D Ellipsoids

Authors: Hannes Stark, Bowen Jing, Tomas Geffner, Jason Yim, Tommi Jaakkola, Arash Vahdat, Karsten Kreis

Abstract: We develop ProtComposer to generate protein structures conditioned on spatial protein layouts that are specified via a set of 3D ellipsoids capturing substructure shapes and semantics. At inference time, we condition on ellipsoids that are hand-constructed, extracted from existing proteins, or from a statistical model, with each option unlocking new capabilities. Hand-specifying ellipsoids enables… ▽ More We develop ProtComposer to generate protein structures conditioned on spatial protein layouts that are specified via a set of 3D ellipsoids capturing substructure shapes and semantics. At inference time, we condition on ellipsoids that are hand-constructed, extracted from existing proteins, or from a statistical model, with each option unlocking new capabilities. Hand-specifying ellipsoids enables users to control the location, size, orientation, secondary structure, and approximate shape of protein substructures. Conditioning on ellipsoids of existing proteins enables redesigning their substructure's connectivity or editing substructure properties. By conditioning on novel and diverse ellipsoid layouts from a simple statistical model, we improve protein generation with expanded Pareto frontiers between designability, novelty, and diversity. Further, this enables sampling designable proteins with a helix-fraction that matches PDB proteins, unlike existing generative models that commonly oversample conceptually simple helix bundles. Code is available at https://github.com/NVlabs/protcomposer. △ Less

Submitted 6 March, 2025; originally announced March 2025.

Comments: Published as a conference paper and Oral (top 1.8%) at ICLR 2025

Emergent collective behavior of cohesive, aligning particles

Authors: Jeanine Shea, Holger Stark

Abstract: Collective behavior is all around us, from flocks of birds to schools of fish. These systems are immensely complex, which makes it pertinent to study their behavior through minimal models. We introduce such a minimal model for cohesive and aligning self-propelled particles in which group cohesion is established through additive, non-reciprocal torques. These torques cause constituents to effective… ▽ More Collective behavior is all around us, from flocks of birds to schools of fish. These systems are immensely complex, which makes it pertinent to study their behavior through minimal models. We introduce such a minimal model for cohesive and aligning self-propelled particles in which group cohesion is established through additive, non-reciprocal torques. These torques cause constituents to effectively turn towards one another. We additionally incorporate an alignment torque, which competes with the cohesive torque in the same spatial range. By changing the strength and range of these torque interactions, we uncover six states which we distinguish via their static and dynamic properties: a disperse state, a multiple worm state, a line state, a persistent worm state, a rotary worm state, and an aster state. Their occurrence strongly depends on initial conditions and stochasticity, so the model exhibits multistabilities. A number of the states exhibit collective dynamics which are reminiscent of those seen in nature. △ Less

Submitted 28 January, 2025; originally announced January 2025.

Trypanosoma brucei moving in microchannels and through constrictions

Authors: Zihan Tan, Julian I. U. Peters, Holger Stark

Abstract: Trypanosoma brucei (T. brucei), a single-celled parasite and natural microswimmer, is responsible for fatal sleeping sickness in infected mammals, including humans. Understanding how T. brucei interacts with fluid environments and navigates through confining spaces is crucial not only for medical and clinical applications but also for a fundamental understanding of how life organizes in a confined… ▽ More Trypanosoma brucei (T. brucei), a single-celled parasite and natural microswimmer, is responsible for fatal sleeping sickness in infected mammals, including humans. Understanding how T. brucei interacts with fluid environments and navigates through confining spaces is crucial not only for medical and clinical applications but also for a fundamental understanding of how life organizes in a confined microscopic world. Using a hybrid multi-particle collision dynamics (MPCD)--molecular dynamics (MD) approach, we present our investigations on the locomotion of an in silico T. brucei in three types of fluid environments: bulk fluid, straight cylindrical microchannels, and microchannels with constrictions. We observe that the helical swimming trajectory of the in silico T. brucei becomes rectified in straight cylindrical channels compared to bulk fluid. The swimming speed for different channel widths is governed by the diameter of the helical trajectory. The speed first slightly increases as the channel narrows and then decreases when the helix diameter is compressed. An optimal swimming speed is achieved, when the channel width is approximately twice the bulk helix diameter. It results from an interplay of the trypanosome's hydrodynamic interactions with the cylindrical channel walls and the high deformability of the parasite. In microchannels with constrictions, the motions of the anterior and posterior ends, the end-to-end distance, and the log-rolling motion of the cell body are characterized and show salient differences compared to the straight-channel case. Depending on the constriction length and width, we observe characteristic slip, stuck, and stuck-slip motions of the model T. brucei within the constriction. Our findings may provide some mechanical insights into how T. brucei moves through blood vessels and tissues, and across the blood-brain barrier. △ Less

Submitted 30 May, 2025; v1 submitted 23 December, 2024; originally announced December 2024.

Comments: Accepted for publication in New J. Phys. (2025), 31 pages, 13 figures

Journal ref: New J. Phys., 2025, 27 064401

Training on test proteins improves fitness, structure, and function prediction

Authors: Anton Bushuiev, Roman Bushuiev, Nikola Zadorozhny, Raman Samusevich, Hannes Stärk, Jiri Sedlar, Tomáš Pluskal, Josef Sivic

Abstract: Data scarcity and distribution shifts often hinder the ability of machine learning models to generalize when applied to proteins and other biological data. Self-supervised pre-training on large datasets is a common method to enhance generalization. However, striving to perform well on all possible proteins can limit model's capacity to excel on any specific one, even though practitioners are often… ▽ More Data scarcity and distribution shifts often hinder the ability of machine learning models to generalize when applied to proteins and other biological data. Self-supervised pre-training on large datasets is a common method to enhance generalization. However, striving to perform well on all possible proteins can limit model's capacity to excel on any specific one, even though practitioners are often most interested in accurate predictions for the individual protein they study. To address this limitation, we propose an orthogonal approach to achieve generalization. Building on the prevalence of self-supervised pre-training, we introduce a method for self-supervised fine-tuning at test time, allowing models to adapt to the test protein of interest on the fly and without requiring any additional data. We study our test-time training (TTT) method through the lens of perplexity minimization and show that it consistently enhances generalization across different models, their scales, and datasets. Notably, our method leads to new state-of-the-art results on the standard benchmark for protein fitness prediction, improves protein structure prediction for challenging targets, and enhances function prediction accuracy. △ Less

Submitted 4 November, 2024; originally announced November 2024.

ET-Flow: Equivariant Flow-Matching for Molecular Conformer Generation

Authors: Majdi Hassan, Nikhil Shenoy, Jungyoon Lee, Hannes Stark, Stephan Thaler, Dominique Beaini

Abstract: Predicting low-energy molecular conformations given a molecular graph is an important but challenging task in computational drug discovery. Existing state-of-the-art approaches either resort to large scale transformer-based models that diffuse over conformer fields, or use computationally expensive methods to generate initial structures and diffuse over torsion angles. In this work, we introduce E… ▽ More Predicting low-energy molecular conformations given a molecular graph is an important but challenging task in computational drug discovery. Existing state-of-the-art approaches either resort to large scale transformer-based models that diffuse over conformer fields, or use computationally expensive methods to generate initial structures and diffuse over torsion angles. In this work, we introduce Equivariant Transformer Flow (ET-Flow). We showcase that a well-designed flow matching approach with equivariance and harmonic prior alleviates the need for complex internal geometry calculations and large architectures, contrary to the prevailing methods in the field. Our approach results in a straightforward and scalable method that directly operates on all-atom coordinates with minimal assumptions. With the advantages of equivariance and flow matching, ET-Flow significantly increases the precision and physical validity of the generated conformers, while being a lighter model and faster at inference. Code is available https://github.com/shenoynikhil/ETFlow. △ Less

Submitted 29 October, 2024; originally announced October 2024.

Comments: NeurIPS 2024

Think While You Generate: Discrete Diffusion with Planned Denoising

Authors: Sulin Liu, Juno Nam, Andrew Campbell, Hannes Stärk, Yilun Xu, Tommi Jaakkola, Rafael Gómez-Bombarelli

Abstract: Discrete diffusion has achieved state-of-the-art performance, outperforming or approaching autoregressive models on standard benchmarks. In this work, we introduce Discrete Diffusion with Planned Denoising (DDPD), a novel framework that separates the generation process into two models: a planner and a denoiser. At inference time, the planner selects which positions to denoise next by identifying t… ▽ More Discrete diffusion has achieved state-of-the-art performance, outperforming or approaching autoregressive models on standard benchmarks. In this work, we introduce Discrete Diffusion with Planned Denoising (DDPD), a novel framework that separates the generation process into two models: a planner and a denoiser. At inference time, the planner selects which positions to denoise next by identifying the most corrupted positions in need of denoising, including both initially corrupted and those requiring additional refinement. This plan-and-denoise approach enables more efficient reconstruction during generation by iteratively identifying and denoising corruptions in the optimal order. DDPD outperforms traditional denoiser-only mask diffusion methods, achieving superior results on language modeling benchmarks such as text8, OpenWebText, and token-based image generation on ImageNet $256 \times 256$. Notably, in language modeling, DDPD significantly reduces the performance gap between diffusion-based and autoregressive methods in terms of generative perplexity. Code is available at https://github.com/liusulin/DDPD. △ Less

Submitted 9 April, 2025; v1 submitted 8 October, 2024; originally announced October 2024.

Comments: ICLR 2025

PyHEP.dev 2024 Workshop Summary Report, August 26-30 2024, Aachen, Germany

Authors: Azzah Alshehri, Jan Bürger, Saransh Chopra, Niclas Eich, Jonas Eppelt, Martin Erdmann, Jonas Eschle, Peter Fackeldey, Maté Farkas, Matthew Feickert, Tristan Fillinger, Benjamin Fischer, Stefan Fröse, Lino Oscar Gerlach, Nikolai Hartmann, Alexander Heidelbach, Alexander Held, Marian I Ivanov, Josué Molina, Yaroslav Nikitenko, Ianna Osborne, Vincenzo Eduardo Padulano, Jim Pivarski, Cyrille Praz, Marcel Rieger , et al. (6 additional authors not shown)

Abstract: The second PyHEP.dev workshop, part of the "Python in HEP Developers" series organized by the HEP Software Foundation (HSF), took place in Aachen, Germany, from August 26 to 30, 2024. This gathering brought together nearly 30 Python package developers, maintainers, and power users to engage in informal discussions about current trends in Python, with a primary focus on analysis tools and technique… ▽ More The second PyHEP.dev workshop, part of the "Python in HEP Developers" series organized by the HEP Software Foundation (HSF), took place in Aachen, Germany, from August 26 to 30, 2024. This gathering brought together nearly 30 Python package developers, maintainers, and power users to engage in informal discussions about current trends in Python, with a primary focus on analysis tools and techniques in High Energy Physics (HEP). The workshop agenda encompassed a range of topics, such as defining the scope of HEP data analysis, exploring the Analysis Grand Challenge project, evaluating statistical models and serialization methods, assessing workflow management systems, examining histogramming practices, and investigating distributed processing tools like RDataFrame, Coffea, and Dask. Additionally, the workshop dedicated time to brainstorming the organization of future PyHEP.dev events, upholding the tradition of alternating between Europe and the United States as host locations. This document, prepared by the session conveners in the weeks following the workshop, serves as a summary of the key discussions, salient points, and conclusions that emerged. △ Less

Submitted 17 December, 2024; v1 submitted 2 October, 2024; originally announced October 2024.

Comments: 16 pages, 0 figures; workshop summary

Generative Modeling of Molecular Dynamics Trajectories

Authors: Bowen Jing, Hannes Stärk, Tommi Jaakkola, Bonnie Berger

Abstract: Molecular dynamics (MD) is a powerful technique for studying microscopic phenomena, but its computational cost has driven significant interest in the development of deep learning-based surrogate models. We introduce generative modeling of molecular trajectories as a paradigm for learning flexible multi-task surrogate models of MD from data. By conditioning on appropriately chosen frames of the tra… ▽ More Molecular dynamics (MD) is a powerful technique for studying microscopic phenomena, but its computational cost has driven significant interest in the development of deep learning-based surrogate models. We introduce generative modeling of molecular trajectories as a paradigm for learning flexible multi-task surrogate models of MD from data. By conditioning on appropriately chosen frames of the trajectory, we show such generative models can be adapted to diverse tasks such as forward simulation, transition path sampling, and trajectory upsampling. By alternatively conditioning on part of the molecular system and inpainting the rest, we also demonstrate the first steps towards dynamics-conditioned molecular design. We validate the full set of these capabilities on tetrapeptide simulations and show that our model can produce reasonable ensembles of protein monomers. Altogether, our work illustrates how generative modeling can unlock value from MD data towards diverse downstream tasks that are not straightforward to address with existing methods or even MD itself. Code is available at https://github.com/bjing2016/mdgen. △ Less

Submitted 26 September, 2024; originally announced September 2024.

Comments: NeurIPS 2024

CodonMPNN for Organism Specific and Codon Optimal Inverse Folding

Authors: Hannes Stark, Umesh Padia, Julia Balla, Cameron Diao, George Church

Abstract: Generating protein sequences conditioned on protein structures is an impactful technique for protein engineering. When synthesizing engineered proteins, they are commonly translated into DNA and expressed in an organism such as yeast. One difficulty in this process is that the expression rates can be low due to suboptimal codon sequences for expressing a protein in a host organism. We propose Codo… ▽ More Generating protein sequences conditioned on protein structures is an impactful technique for protein engineering. When synthesizing engineered proteins, they are commonly translated into DNA and expressed in an organism such as yeast. One difficulty in this process is that the expression rates can be low due to suboptimal codon sequences for expressing a protein in a host organism. We propose CodonMPNN, which generates a codon sequence conditioned on a protein backbone structure and an organism label. If naturally occurring DNA sequences are close to codon optimality, CodonMPNN could learn to generate codon sequences with higher expression yields than heuristic codon choices for generated amino acid sequences. Experiments show that CodonMPNN retains the performance of previous inverse folding approaches and recovers wild-type codons more frequently than baselines. Furthermore, CodonMPNN has a higher likelihood of generating high-fitness codon sequences than low-fitness codon sequences for the same protein sequence. Code is available at https://github.com/HannesStark/CodonMPNN. △ Less

Submitted 25 September, 2024; originally announced September 2024.

Comments: Appeared at the 2024 ICML AI4Science workshop

Mixing in viscoelastic fluids using elastic turbulence

Authors: Reinier van Buel, Holger Stark

Abstract: We investigate the influence of elastic turbulence on mixing of a scalar concentration field within a viscoelastic fluid in a two-dimensional Taylor-Couette geometry using numerical solutions of the Oldroyd-B model. The flow state is determined through the secondary-flow order parameter indicating the regime of elastic turbulence. When starting in the turbulent state and subsequently lowering the… ▽ More We investigate the influence of elastic turbulence on mixing of a scalar concentration field within a viscoelastic fluid in a two-dimensional Taylor-Couette geometry using numerical solutions of the Oldroyd-B model. The flow state is determined through the secondary-flow order parameter indicating the regime of elastic turbulence. When starting in the turbulent state and subsequently lowering the Weissenberg number, a weakly-chaotic flow occurs below $\mathrm{Wi_c}$. Advection in both the turbulent and weakly-chaotic flow states induces mixing, which we illustrate by the time evolution of the standard deviation of the solute concentration from the uniform distribution. In particular, in the elastic turbulent state mixing is strong and we quantify it by the mixing rate, the mixing time, and the mixing efficiency. All three quantities follow scaling laws. Importantly, we show that the order parameter is strongly correlated to the mixing rate and hence is also a good indication of mixing within the fluid. △ Less

Submitted 1 January, 2025; v1 submitted 10 September, 2024; originally announced September 2024.

A Page-Rank-like Approach to Optimal Placement of Charging Stations in a Warehouse

Authors: Hans-Georg Stark, Mustafa Jelibaghu, Katrin Tschirpke, Michael Eley

Abstract: In this paper we describe an approach to the problem of finding optimal positions for charging stations (CS) in a warehouse, where a fleet of electrical industrial trucks/forklifts is employed. Our procedure is motivated by Googles Page-Ranking. The graph model underlying our method is easily extensible from simple distance based criteria, relevant for choosing optimal CS-positions, to more comple… ▽ More In this paper we describe an approach to the problem of finding optimal positions for charging stations (CS) in a warehouse, where a fleet of electrical industrial trucks/forklifts is employed. Our procedure is motivated by Googles Page-Ranking. The graph model underlying our method is easily extensible from simple distance based criteria, relevant for choosing optimal CS-positions, to more complex criteria taking into account, e.g., the ''state of charge'' (SOC) of the individual trucks battery. △ Less

Submitted 24 June, 2024; originally announced June 2024.

CowScape: Quantitative reconstruction of the conformational landscape of biological macromolecules from cryo-EM data

Authors: Felix Lambrecht, Andreas Kröpelin, Mario Lüttich, Michael Habeck, David Haselbach, Holger Stark

Abstract: Cryo-EM data processing typically focuses on the structure of the main conformational state under investigation and discards images that belong to other states. This approach can reach atomic resolution, but ignores vast amounts of valuable information about the underlying conformational ensemble and its dynamics. CowScape analyzes an entire cryo-EM dataset and thereby obtains a quantitative descr… ▽ More Cryo-EM data processing typically focuses on the structure of the main conformational state under investigation and discards images that belong to other states. This approach can reach atomic resolution, but ignores vast amounts of valuable information about the underlying conformational ensemble and its dynamics. CowScape analyzes an entire cryo-EM dataset and thereby obtains a quantitative description of structural variability of macromolecular complexes that represents the biochemically relevant conformational space. By combining extensive image classification with principal component analysis (PCA) of the classified 3D volumes and kernel density estimation, CowScape can be used as a quantitative tool to analyze this variability. PCA projects all 3D structures along the major modes spanning a low-dimensional space that captures a large portion of structural variability. The number of particle images in a given state can be used to calculate an energy landscape based on kernel density estimation and Boltzmann inversion. By revealing allosteric interactions in macromolecular complexes, CowScape allows us to distinguish and interpret dynamic changes in macromolecular complexes during function and regulation. △ Less

Submitted 18 February, 2024; originally announced February 2024.

Comments: 15 pages + 4 figures (main text)

Dirichlet Flow Matching with Applications to DNA Sequence Design

Authors: Hannes Stark, Bowen Jing, Chenyu Wang, Gabriele Corso, Bonnie Berger, Regina Barzilay, Tommi Jaakkola

Abstract: Discrete diffusion or flow models could enable faster and more controllable sequence generation than autoregressive models. We show that naïve linear flow matching on the simplex is insufficient toward this goal since it suffers from discontinuities in the training target and further pathologies. To overcome this, we develop Dirichlet flow matching on the simplex based on mixtures of Dirichlet dis… ▽ More Discrete diffusion or flow models could enable faster and more controllable sequence generation than autoregressive models. We show that naïve linear flow matching on the simplex is insufficient toward this goal since it suffers from discontinuities in the training target and further pathologies. To overcome this, we develop Dirichlet flow matching on the simplex based on mixtures of Dirichlet distributions as probability paths. In this framework, we derive a connection between the mixtures' scores and the flow's vector field that allows for classifier and classifier-free guidance. Further, we provide distilled Dirichlet flow matching, which enables one-step sequence generation with minimal performance hits, resulting in $O(L)$ speedups compared to autoregressive models. On complex DNA sequence generation tasks, we demonstrate superior performance compared to all baselines in distributional metrics and in achieving desired design targets for generated sequences. Finally, we show that our classifier-free guidance approach improves unconditional generation and is effective for generating DNA that satisfies design targets. Code is available at https://github.com/HannesStark/dirichlet-flow-matching. △ Less

Submitted 30 May, 2024; v1 submitted 8 February, 2024; originally announced February 2024.

Comments: Published at ICML 2024. (Proceedings of the 41st International Conference on Machine Learning, Vienna, Austria. PMLR 235, 2024)

Transition Path Sampling with Boltzmann Generator-based MCMC Moves

Authors: Michael Plainer, Hannes Stärk, Charlotte Bunne, Stephan Günnemann

Abstract: Sampling all possible transition paths between two 3D states of a molecular system has various applications ranging from catalyst design to drug discovery. Current approaches to sample transition paths use Markov chain Monte Carlo and rely on time-intensive molecular dynamics simulations to find new paths. Our approach operates in the latent space of a normalizing flow that maps from the molecule'… ▽ More Sampling all possible transition paths between two 3D states of a molecular system has various applications ranging from catalyst design to drug discovery. Current approaches to sample transition paths use Markov chain Monte Carlo and rely on time-intensive molecular dynamics simulations to find new paths. Our approach operates in the latent space of a normalizing flow that maps from the molecule's Boltzmann distribution to a Gaussian, where we propose new paths without requiring molecular simulations. Using alanine dipeptide, we explore Metropolis-Hastings acceptance criteria in the latent space for exact sampling and investigate different latent proposal mechanisms. △ Less

Submitted 28 May, 2024; v1 submitted 8 December, 2023; originally announced December 2023.

Comments: Spotlight at NeurIPS 2023 Generative AI and Biology Workshop

Steering droplets on substrates with plane-wave wettability patterns and deformations

Authors: Josua Grawitter, Holger Stark

Abstract: Motivated by strategies for targeted microfluidic transport of droplets, we investigate how sessile droplets can be steered toward a preferred direction using travelling waves in substrate wettability or deformations of the substrate. To perform our numerical study, we implement the boundary-element method to solve the governing Stokes equations for the fluid flow field inside the moving droplet.… ▽ More Motivated by strategies for targeted microfluidic transport of droplets, we investigate how sessile droplets can be steered toward a preferred direction using travelling waves in substrate wettability or deformations of the substrate. To perform our numerical study, we implement the boundary-element method to solve the governing Stokes equations for the fluid flow field inside the moving droplet. In both cases we find two distinct modes of droplet motion. For small wave speed the droplet surfes with a constant velocity on the wave, while beyond a critical wave speed a periodic wobbling motion occurs, the period of which diverges at the transition. These observation can be rationalized by the nonuniform oscillator model and the transition described by a SNIPER bifurcation. For the travelling waves in wettability the mean droplet velocity in the wobbling state decays with the inverse wave speed. In contrast, for travelling-wave deformations of the substrate it is proportional to the wave speed at large speed values since the droplet always has to move up and down. To rationalize this behavior, the nonuniform oscillator model has to be extended. Since the critical wave speed of the bifurcation depends on the droplet radius, this dependence can be used to sort droplets by size. △ Less

Submitted 30 October, 2023; originally announced October 2023.

Comments: 13 pages, 5 figures

Journal ref: Soft Matter 20 (2024) 3161-3174

Harmonic Self-Conditioned Flow Matching for Multi-Ligand Docking and Binding Site Design

Authors: Hannes Stärk, Bowen Jing, Regina Barzilay, Tommi Jaakkola

Abstract: A significant amount of protein function requires binding small molecules, including enzymatic catalysis. As such, designing binding pockets for small molecules has several impactful applications ranging from drug synthesis to energy storage. Towards this goal, we first develop HarmonicFlow, an improved generative process over 3D protein-ligand binding structures based on our self-conditioned flow… ▽ More A significant amount of protein function requires binding small molecules, including enzymatic catalysis. As such, designing binding pockets for small molecules has several impactful applications ranging from drug synthesis to energy storage. Towards this goal, we first develop HarmonicFlow, an improved generative process over 3D protein-ligand binding structures based on our self-conditioned flow matching objective. FlowSite extends this flow model to jointly generate a protein pocket's discrete residue types and the molecule's binding 3D structure. We show that HarmonicFlow improves upon state-of-the-art generative processes for docking in simplicity, generality, and average sample quality in pocket-level docking. Enabled by this structure modeling, FlowSite designs binding sites substantially better than baseline approaches. △ Less

Submitted 30 May, 2024; v1 submitted 9 October, 2023; originally announced October 2023.

Comments: Published at ICML 2024. (Proceedings of the 41st International Conference on Machine Learning, Vienna, Austria. PMLR 235, 2024)

Dividing active and passive particles in nonuniform nutrient environments

Authors: Till Welker, Holger Stark

Abstract: To explore the coupling between a growing population of microorganisms such as E. coli and a nonuniform nutrient distribution, we formulate a minimalistic model. It consists of active Brownian particles that divide and grow at a nutrient-dependent rate following the Monod equation. The nutrient concentration obeys a diffusion equation with a consumption term and a point source. In this setting the… ▽ More To explore the coupling between a growing population of microorganisms such as E. coli and a nonuniform nutrient distribution, we formulate a minimalistic model. It consists of active Brownian particles that divide and grow at a nutrient-dependent rate following the Monod equation. The nutrient concentration obeys a diffusion equation with a consumption term and a point source. In this setting the heterogeneity in the nutrient distribution can be tuned by the diffusion coefficient. In particle-based simulations, we demonstrate that passive and weakly active particles form proliferation-induced clusters when the nutrient is localized, without relying on further mechanisms such as chemotaxis or adhesion. In contrast, strongly active particles disperse in the whole system during their lifetime and no clustering is present. The steady population is unaffected by activity or nonuniform nutrient distribution and only determined by the ratio of nutrient influx and bacterial death. However, the transient dynamics strongly depends on the nutrient distribution and activity. Passive particles in almost uniform nutrient profiles display a strong population overshoot, with clusters forming all over the system. In contrast, when slowly diffusing nutrients remain centred around the source, the bacterial population quickly approaches the steady state due to its strong coupling to the nutrient. Conversely, the population overshoot of highly active particles becomes stronger when the nutrient localisation increases. We successfully map the transient population dynamics onto a uniform model where the effect of the nonuniform nutrient and bacterial distributions are rationalized by two effective areas. △ Less

Submitted 10 January, 2024; v1 submitted 2 October, 2023; originally announced October 2023.

Comments: 13 figures

Artificial Intelligence for Science in Quantum, Atomistic, and Continuum Systems

Authors: Xuan Zhang, Limei Wang, Jacob Helwig, Youzhi Luo, Cong Fu, Yaochen Xie, Meng Liu, Yuchao Lin, Zhao Xu, Keqiang Yan, Keir Adams, Maurice Weiler, Xiner Li, Tianfan Fu, Yucheng Wang, Alex Strasser, Haiyang Yu, YuQing Xie, Xiang Fu, Shenglong Xu, Yi Liu, Yuanqi Du, Alexandra Saxton, Hongyi Ling, Hannah Lawrence , et al. (38 additional authors not shown)

Abstract: Advances in artificial intelligence (AI) are fueling a new paradigm of discoveries in natural sciences. Today, AI has started to advance natural sciences by improving, accelerating, and enabling our understanding of natural phenomena at a wide range of spatial and temporal scales, giving rise to a new area of research known as AI for science (AI4Science). Being an emerging research paradigm, AI4Sc… ▽ More Advances in artificial intelligence (AI) are fueling a new paradigm of discoveries in natural sciences. Today, AI has started to advance natural sciences by improving, accelerating, and enabling our understanding of natural phenomena at a wide range of spatial and temporal scales, giving rise to a new area of research known as AI for science (AI4Science). Being an emerging research paradigm, AI4Science is unique in that it is an enormous and highly interdisciplinary area. Thus, a unified and technical treatment of this field is needed yet challenging. This work aims to provide a technically thorough account of a subarea of AI4Science; namely, AI for quantum, atomistic, and continuum systems. These areas aim at understanding the physical world from the subatomic (wavefunctions and electron density), atomic (molecules, proteins, materials, and interactions), to macro (fluids, climate, and subsurface) scales and form an important subarea of AI4Science. A unique advantage of focusing on these areas is that they largely share a common set of challenges, thereby allowing a unified and foundational treatment. A key common challenge is how to capture physics first principles, especially symmetries, in natural systems by deep learning methods. We provide an in-depth yet intuitive account of techniques to achieve equivariance to symmetry transformations. We also discuss other common technical challenges, including explainability, out-of-distribution generalization, knowledge transfer with foundation and large language models, and uncertainty quantification. To facilitate learning and education, we provide categorized lists of resources that we found to be useful. We strive to be thorough and unified and hope this initial effort may trigger more community interests and efforts to further advance AI4Science. △ Less

Submitted 10 June, 2025; v1 submitted 17 July, 2023; originally announced July 2023.

Comments: Accepted to Foundations and Trends in Machine Learning

DiffDock-PP: Rigid Protein-Protein Docking with Diffusion Models

Authors: Mohamed Amine Ketata, Cedrik Laue, Ruslan Mammadov, Hannes Stärk, Menghua Wu, Gabriele Corso, Céline Marquet, Regina Barzilay, Tommi S. Jaakkola

Abstract: Understanding how proteins structurally interact is crucial to modern biology, with applications in drug discovery and protein design. Recent machine learning methods have formulated protein-small molecule docking as a generative problem with significant performance boosts over both traditional and deep learning baselines. In this work, we propose a similar approach for rigid protein-protein docki… ▽ More Understanding how proteins structurally interact is crucial to modern biology, with applications in drug discovery and protein design. Recent machine learning methods have formulated protein-small molecule docking as a generative problem with significant performance boosts over both traditional and deep learning baselines. In this work, we propose a similar approach for rigid protein-protein docking: DiffDock-PP is a diffusion generative model that learns to translate and rotate unbound protein structures into their bound conformations. We achieve state-of-the-art performance on DIPS with a median C-RMSD of 4.85, outperforming all considered baselines. Additionally, DiffDock-PP is faster than all search-based methods and generates reliable confidence estimates for its predictions. Our code is publicly available at $\texttt{https://github.com/ketatam/DiffDock-PP}$ △ Less

Submitted 7 April, 2023; originally announced April 2023.

Comments: ICLR Machine Learning for Drug Discovery (MLDD) Workshop 2023

Orientational dynamics and rheology of active suspensions in weakly viscoelastic flows

Authors: Akash Choudhary, Sankalp Nambiar, Holger Stark

Abstract: Microswimmer suspensions in Newtonian fluids exhibit unusual macroscale properties, such as a superfluidic behavior, which can be harnessed to perform work at microscopic scales. Since most biological fluids are non-Newtonian, here we study the rheology of a microswimmer suspension in a weakly viscoelastic shear flow. At the individual level, we find that the viscoelastic stresses generated by act… ▽ More Microswimmer suspensions in Newtonian fluids exhibit unusual macroscale properties, such as a superfluidic behavior, which can be harnessed to perform work at microscopic scales. Since most biological fluids are non-Newtonian, here we study the rheology of a microswimmer suspension in a weakly viscoelastic shear flow. At the individual level, we find that the viscoelastic stresses generated by activity substantially modify the Jeffery orbits well-known from Newtonian fluids. The orientational dynamics depends on the swimmer type; especially pushers can resist flow-induced rotation and align at an angle with the flow. To analyze its impact on bulk rheology, we study a dilute microswimmer suspension in the presence of random tumbling and rotational diffusion. Strikingly, swimmer activity and its elastic response in polymeric fluids alter the orientational distribution and substantially amplify the swimmer-induced viscosity. This suggests that pusher suspensions reach the superfluidic regime at lower volume fractions compared to a Newtonian fluid with identical viscosity. △ Less

Submitted 15 September, 2023; v1 submitted 27 March, 2023; originally announced March 2023.

Comments: Communications Physics 6, 163 (2023)

Towards a Muon Collider

Authors: Carlotta Accettura, Dean Adams, Rohit Agarwal, Claudia Ahdida, Chiara Aimè, Nicola Amapane, David Amorim, Paolo Andreetto, Fabio Anulli, Robert Appleby, Artur Apresyan, Aram Apyan, Sergey Arsenyev, Pouya Asadi, Mohammed Attia Mahmoud, Aleksandr Azatov, John Back, Lorenzo Balconi, Laura Bandiera, Roger Barlow, Nazar Bartosik, Emanuela Barzi, Fabian Batsch, Matteo Bauce, J. Scott Berg , et al. (272 additional authors not shown)

Abstract: A muon collider would enable the big jump ahead in energy reach that is needed for a fruitful exploration of fundamental interactions. The challenges of producing muon collisions at high luminosity and 10 TeV centre of mass energy are being investigated by the recently-formed International Muon Collider Collaboration. This Review summarises the status and the recent advances on muon colliders desi… ▽ More A muon collider would enable the big jump ahead in energy reach that is needed for a fruitful exploration of fundamental interactions. The challenges of producing muon collisions at high luminosity and 10 TeV centre of mass energy are being investigated by the recently-formed International Muon Collider Collaboration. This Review summarises the status and the recent advances on muon colliders design, physics and detector studies. The aim is to provide a global perspective of the field and to outline directions for future work. △ Less

Submitted 27 November, 2023; v1 submitted 15 March, 2023; originally announced March 2023.

Comments: 118 pages, 103 figures

Task-Agnostic Graph Neural Network Evaluation via Adversarial Collaboration

Authors: Xiangyu Zhao, Hannes Stärk, Dominique Beaini, Yiren Zhao, Pietro Liò

Abstract: It has been increasingly demanding to develop reliable methods to evaluate the progress of Graph Neural Network (GNN) research for molecular representation learning. Existing GNN benchmarking methods for molecular representation learning focus on comparing the GNNs' performances on some node/graph classification/regression tasks on certain datasets. However, there lacks a principled, task-agnostic… ▽ More It has been increasingly demanding to develop reliable methods to evaluate the progress of Graph Neural Network (GNN) research for molecular representation learning. Existing GNN benchmarking methods for molecular representation learning focus on comparing the GNNs' performances on some node/graph classification/regression tasks on certain datasets. However, there lacks a principled, task-agnostic method to directly compare two GNNs. Additionally, most of the existing self-supervised learning works incorporate handcrafted augmentations to the data, which has several severe difficulties to be applied on graphs due to their unique characteristics. To address the aforementioned issues, we propose GraphAC (Graph Adversarial Collaboration) -- a conceptually novel, principled, task-agnostic, and stable framework for evaluating GNNs through contrastive self-supervision. We introduce a novel objective function: the Competitive Barlow Twins, that allow two GNNs to jointly update themselves from direct competitions against each other. GraphAC succeeds in distinguishing GNNs of different expressiveness across various aspects, and has demonstrated to be a principled and reliable GNN evaluation method, without necessitating any augmentations. △ Less

Submitted 26 March, 2023; v1 submitted 26 January, 2023; originally announced January 2023.

Comments: 11th International Conference on Learning Representations (ICLR 2023) Machine Learning for Drug Discovery (MLDD) Workshop. 17 pages, 6 figures, 4 tables

Generalized Laplacian Positional Encoding for Graph Representation Learning

Authors: Sohir Maskey, Ali Parviz, Maximilian Thiessen, Hannes Stärk, Ylli Sadikaj, Haggai Maron

Abstract: Graph neural networks (GNNs) are the primary tool for processing graph-structured data. Unfortunately, the most commonly used GNNs, called Message Passing Neural Networks (MPNNs) suffer from several fundamental limitations. To overcome these limitations, recent works have adapted the idea of positional encodings to graph data. This paper draws inspiration from the recent success of Laplacian-based… ▽ More Graph neural networks (GNNs) are the primary tool for processing graph-structured data. Unfortunately, the most commonly used GNNs, called Message Passing Neural Networks (MPNNs) suffer from several fundamental limitations. To overcome these limitations, recent works have adapted the idea of positional encodings to graph data. This paper draws inspiration from the recent success of Laplacian-based positional encoding and defines a novel family of positional encoding schemes for graphs. We accomplish this by generalizing the optimization problem that defines the Laplace embedding to more general dissimilarity functions rather than the 2-norm used in the original formulation. This family of positional encodings is then instantiated by considering p-norms. We discuss a method for calculating these positional encoding schemes, implement it in PyTorch and demonstrate how the resulting positional encoding captures different properties of the graph. Furthermore, we demonstrate that this novel family of positional encodings can improve the expressive power of MPNNs. Lastly, we present preliminary experimental results. △ Less

Submitted 10 November, 2022; v1 submitted 28 October, 2022; originally announced October 2022.

Comments: Accepted at NeurIPS Workshop on Symmetry and Geometry in Neural Representations: Extended Abstract Track 2022

Equivariant 3D-Conditional Diffusion Models for Molecular Linker Design

Authors: Ilia Igashov, Hannes Stärk, Clément Vignac, Victor Garcia Satorras, Pascal Frossard, Max Welling, Michael Bronstein, Bruno Correia

Abstract: Fragment-based drug discovery has been an effective paradigm in early-stage drug development. An open challenge in this area is designing linkers between disconnected molecular fragments of interest to obtain chemically-relevant candidate drug molecules. In this work, we propose DiffLinker, an E(3)-equivariant 3D-conditional diffusion model for molecular linker design. Given a set of disconnected… ▽ More Fragment-based drug discovery has been an effective paradigm in early-stage drug development. An open challenge in this area is designing linkers between disconnected molecular fragments of interest to obtain chemically-relevant candidate drug molecules. In this work, we propose DiffLinker, an E(3)-equivariant 3D-conditional diffusion model for molecular linker design. Given a set of disconnected fragments, our model places missing atoms in between and designs a molecule incorporating all the initial fragments. Unlike previous approaches that are only able to connect pairs of molecular fragments, our method can link an arbitrary number of fragments. Additionally, the model automatically determines the number of atoms in the linker and its attachment points to the input fragments. We demonstrate that DiffLinker outperforms other methods on the standard datasets generating more diverse and synthetically-accessible molecules. Besides, we experimentally test our method in real-world applications, showing that it can successfully generate valid linkers conditioned on target protein pockets. △ Less

Submitted 11 October, 2022; originally announced October 2022.

Comments: Under review

DiffDock: Diffusion Steps, Twists, and Turns for Molecular Docking

Authors: Gabriele Corso, Hannes Stärk, Bowen Jing, Regina Barzilay, Tommi Jaakkola

Abstract: Predicting the binding structure of a small molecule ligand to a protein -- a task known as molecular docking -- is critical to drug design. Recent deep learning methods that treat docking as a regression problem have decreased runtime compared to traditional search-based methods but have yet to offer substantial improvements in accuracy. We instead frame molecular docking as a generative modeling… ▽ More Predicting the binding structure of a small molecule ligand to a protein -- a task known as molecular docking -- is critical to drug design. Recent deep learning methods that treat docking as a regression problem have decreased runtime compared to traditional search-based methods but have yet to offer substantial improvements in accuracy. We instead frame molecular docking as a generative modeling problem and develop DiffDock, a diffusion generative model over the non-Euclidean manifold of ligand poses. To do so, we map this manifold to the product space of the degrees of freedom (translational, rotational, and torsional) involved in docking and develop an efficient diffusion process on this space. Empirically, DiffDock obtains a 38% top-1 success rate (RMSD<2A) on PDBBind, significantly outperforming the previous state-of-the-art of traditional docking (23%) and deep learning (20%) methods. Moreover, while previous methods are not able to dock on computationally folded structures (maximum accuracy 10.4%), DiffDock maintains significantly higher precision (21.7%). Finally, DiffDock has fast inference times and provides confidence estimates with high selective accuracy. △ Less

Submitted 11 February, 2023; v1 submitted 4 October, 2022; originally announced October 2022.

Comments: International Conference on Learning Representations (ICLR 2023)

Graph Anisotropic Diffusion

Authors: Ahmed A. A. Elhag, Gabriele Corso, Hannes Stärk, Michael M. Bronstein

Abstract: Traditional Graph Neural Networks (GNNs) rely on message passing, which amounts to permutation-invariant local aggregation of neighbour features. Such a process is isotropic and there is no notion of `direction' on the graph. We present a new GNN architecture called Graph Anisotropic Diffusion. Our model alternates between linear diffusion, for which a closed-form solution is available, and local… ▽ More Traditional Graph Neural Networks (GNNs) rely on message passing, which amounts to permutation-invariant local aggregation of neighbour features. Such a process is isotropic and there is no notion of `direction' on the graph. We present a new GNN architecture called Graph Anisotropic Diffusion. Our model alternates between linear diffusion, for which a closed-form solution is available, and local anisotropic filters to obtain efficient multi-hop anisotropic kernels. We test our model on two common molecular property prediction benchmarks (ZINC and QM9) and show its competitive performance. △ Less

Submitted 30 April, 2022; originally announced May 2022.

Comments: 10 pages, 3 figures, Published at the GTRL and MLDD workshops, ICLR 2022

Collective motion of active particles exhibiting non-reciprocal orientational interactions

Authors: Milos Knezevic, Till Welker, Holger Stark

Abstract: We present a Brownian dynamics study of a 2d bath of active particles interacting among each other through usual steric interactions and, additionally, via non-reciprocal avoidant orientational interactions. We motivate them by the fact that the two flagella of the alga Chlamydomonas interact sterically with nearby surfaces such that a torque acts on the alga. As expected, in most cases such inter… ▽ More We present a Brownian dynamics study of a 2d bath of active particles interacting among each other through usual steric interactions and, additionally, via non-reciprocal avoidant orientational interactions. We motivate them by the fact that the two flagella of the alga Chlamydomonas interact sterically with nearby surfaces such that a torque acts on the alga. As expected, in most cases such interactions disrupt the motility-induced particle clustering in active baths. Surprisingly, however, we find that the active particles can self-organize into collectively moving flocks if the range of non-reciprocal interactions is close to that of steric interactions. We observe that the flocking motion can manifest itself through a variety of structural forms, spanning from single dense bands to multiple moderately-dense stripes, which are highly dynamic. The flocking order parameter is found to be only weakly dependent on the underlying flock structure. Together with the variance of the local-density distribution, one can clearly group the flocking motion into the two separate band and dynamic-stripes states. △ Less

Submitted 20 October, 2022; v1 submitted 12 April, 2022; originally announced April 2022.

Comments: 10 pages, 6 figures

Szilard engines and information-based work extraction for active systems

Authors: P. Malgaretti, H. Stark

Abstract: The out of equilibrium nature of active systems can be exploited for the design of information-based engines. We design two types of an active Szilard engine that use a Maxwell daemon to extract work from an active bath composed of non-interacting Active Brownian Particles (ABPs). The two engines exploit either the quasi-static active pressure of ABPs or the long correlation time of their velociti… ▽ More The out of equilibrium nature of active systems can be exploited for the design of information-based engines. We design two types of an active Szilard engine that use a Maxwell daemon to extract work from an active bath composed of non-interacting Active Brownian Particles (ABPs). The two engines exploit either the quasi-static active pressure of ABPs or the long correlation time of their velocities. For both engines the Landauer principle can be violated so that larger work can be extracted compared to conventional Szilard engines operating in quasi thermal equilibrium. For both of our engines,we identify the optimal regimes at which the work extracted and the efficiency are maximized. Finally, we we discuss them in the context of synthetic and biological active systems. △ Less

Submitted 25 August, 2022; v1 submitted 24 March, 2022; originally announced March 2022.

The physics case of a 3 TeV muon collider stage

Authors: Jorge De Blas, Dario Buttazzo, Rodolfo Capdevilla, David Curtin, Roberto Franceschini, Fabio Maltoni, Patrick Meade, Federico Meloni, Shufang Su, Eleni Vryonidou, Andrea Wulzer, Chiara Aimè, Aram Apyan, Pouya Asadi, Mohammed Attia Mahmoud, Aleksandr Azatov, Nazar Bartosik, Alessandro Bertolin, Salvatore Bottaro, Laura Buonincontri, Massimo Casarsa, Luca Castelli, Maria Gabriella Catanesi, Francesco Giovanni Celiberto, Alessandro Cerri , et al. (109 additional authors not shown)

Abstract: In the path towards a muon collider with center of mass energy of 10 TeV or more, a stage at 3 TeV emerges as an appealing option. Reviewing the physics potential of such muon collider is the main purpose of this document. In order to outline the progression of the physics performances across the stages, a few sensitivity projections for higher energy are also presented. There are many opportuniti… ▽ More In the path towards a muon collider with center of mass energy of 10 TeV or more, a stage at 3 TeV emerges as an appealing option. Reviewing the physics potential of such muon collider is the main purpose of this document. In order to outline the progression of the physics performances across the stages, a few sensitivity projections for higher energy are also presented. There are many opportunities for probing new physics at a 3 TeV muon collider. Some of them are in common with the extensively documented physics case of the CLIC 3 TeV energy stage, and include measuring the Higgs trilinear coupling and testing the possible composite nature of the Higgs boson and of the top quark at the 20 TeV scale. Other opportunities are unique of a 3 TeV muon collider, and stem from the fact that muons are collided rather than electrons. This is exemplified by studying the potential to explore the microscopic origin of the current $g$-2 and $B$-physics anomalies, which are both related with muons. △ Less

Submitted 27 May, 2022; v1 submitted 14 March, 2022; originally announced March 2022.

Comments: 73 pages, 28 figures; Contribution to Snowmass 2021

Muon Collider Physics Summary

Authors: Chiara Aimè, Aram Apyan, Mohammed Attia Mahmoud, Nazar Bartosik, Alessandro Bertolin, Maurizio Bonesini, Salvatore Bottaro, Dario Buttazzo, Rodolfo Capdevilla, Massimo Casarsa, Luca Castelli, Maria Gabriella Catanesi, Francesco Giovanni Celiberto, Alessandro Cerri, Cari Cesarotti, Grigorios Chachamis, Siyu Chen, Yang-Ting Chien, Mauro Chiesa, Gianmaria Collazuol, Marco Costa, Nathaniel Craig, David Curtin, Sridhara Dasu, Jorge De Blas , et al. (100 additional authors not shown)

Abstract: The perspective of designing muon colliders with high energy and luminosity, which is being investigated by the International Muon Collider Collaboration, has triggered a growing interest in their physics reach. We present a concise summary of the muon colliders potential to explore new physics, leveraging on the unique possibility of combining high available energy with very precise measurements. The perspective of designing muon colliders with high energy and luminosity, which is being investigated by the International Muon Collider Collaboration, has triggered a growing interest in their physics reach. We present a concise summary of the muon colliders potential to explore new physics, leveraging on the unique possibility of combining high available energy with very precise measurements. △ Less

Submitted 27 May, 2022; v1 submitted 14 March, 2022; originally announced March 2022.

Comments: 21 pages, 7 figures; Contribution to Snowmass 2021

Characterizing elastic turbulence in the three-dimensional von Karman swirling flow using the Oldroyd-B model

Authors: Reinier van Buel, Holger Stark

Abstract: We present the full three-dimensional numerical investigation of the von Karman swirling flow between two parallel plates using the Oldroyd-B model and characterize the onset and development of elastic turbulence. We quantify the flow state with the secondary-flow strength, a measure of the average strength of the velocity fluctuations, and then define an order parameter as the time average of the… ▽ More We present the full three-dimensional numerical investigation of the von Karman swirling flow between two parallel plates using the Oldroyd-B model and characterize the onset and development of elastic turbulence. We quantify the flow state with the secondary-flow strength, a measure of the average strength of the velocity fluctuations, and then define an order parameter as the time average of the secondary-flow strength. The order parameter displays a subcritical transition from the laminar to a bistable flow that switches between weakly chaotic flow and elastic turbulence. The transition to the bistable flow occurs at the critical Weissenberg number $\mathrm{Wi_c}=12$. Above $\mathrm{Wi_c}$, in the elastic turbulent state, we observe a strong increase in velocity fluctuations and flow resistance, which we define as the total work performed on the fluid. Upon starting simulations in the turbulent state and subsequently lowering $\mathrm{Wi}$ below its critical value, we observe hysteretic behavior in the order parameter and the flow resistance, which is a common feature of a subcritical transition. Hysteresis has also been found in experiments. Additionally, we find power-law scaling in the spatial and temporal power spectra of the velocity fluctuations, characteristic for elastic turbulence. The maximum values of the power-law exponents in our simulations are $α_t= 3.41$ for the temporal exponent and $α_s= 3.17$ for the spatial exponent, which are remarkably close to the values obtained in experiments. △ Less

Submitted 25 February, 2022; originally announced February 2022.

EquiBind: Geometric Deep Learning for Drug Binding Structure Prediction

Authors: Hannes Stärk, Octavian-Eugen Ganea, Lagnajit Pattanaik, Regina Barzilay, Tommi Jaakkola

Abstract: Predicting how a drug-like molecule binds to a specific protein target is a core problem in drug discovery. An extremely fast computational binding method would enable key applications such as fast virtual screening or drug engineering. Existing methods are computationally expensive as they rely on heavy candidate sampling coupled with scoring, ranking, and fine-tuning steps. We challenge this par… ▽ More Predicting how a drug-like molecule binds to a specific protein target is a core problem in drug discovery. An extremely fast computational binding method would enable key applications such as fast virtual screening or drug engineering. Existing methods are computationally expensive as they rely on heavy candidate sampling coupled with scoring, ranking, and fine-tuning steps. We challenge this paradigm with EquiBind, an SE(3)-equivariant geometric deep learning model performing direct-shot prediction of both i) the receptor binding location (blind docking) and ii) the ligand's bound pose and orientation. EquiBind achieves significant speed-ups and better quality compared to traditional and recent baselines. Further, we show extra improvements when coupling it with existing fine-tuning techniques at the cost of increased running time. Finally, we propose a novel and fast fine-tuning model that adjusts torsion angles of a ligand's rotatable bonds based on closed-form global minima of the von Mises angular distance to a given input atomic point cloud, avoiding previous expensive differential evolution strategies for energy minimization. △ Less

Submitted 4 June, 2022; v1 submitted 7 February, 2022; originally announced February 2022.

Comments: 39th International Conference on Machine Learning (ICML 2022). Also accepted at ICLR 2022 GTRL and at ICLR 2022 MLDD as spotlight

Journal ref: 39th International Conference on Machine Learning (ICML 2022)

Gyrotactic cluster formation of bottom-heavy squirmers

Authors: Felix Rühle, Arne W. Zantop, Holger Stark

Abstract: Squirmers that are bottom-heavy experience a torque that aligns them along the vertical so that they swim upwards. In a suspension of many squirmers, they also interact hydrodynamically via flow fields that are initiated by their swimming motion and by gravity. Swimming under the combined action of flow field vorticity and gravitational torque is called gyrotaxis. Using the method of multi-particl… ▽ More Squirmers that are bottom-heavy experience a torque that aligns them along the vertical so that they swim upwards. In a suspension of many squirmers, they also interact hydrodynamically via flow fields that are initiated by their swimming motion and by gravity. Swimming under the combined action of flow field vorticity and gravitational torque is called gyrotaxis. Using the method of multi-particle collision dynamics, we perform hydrodynamic simulations of a many-squirmer system floating above the bottom surface. Due to gyrotaxis they exhibit pronounced cluster formation with increasing gravitational torque. The clusters are more volatile at low values but compactify to smaller clusters at larger torques. The mean distance between clusters is mainly controlled by the gravitational torque and not the global density. Furthermore, we observe that neutral squirmers form clusters more easily, whereas pullers require larger gravitational torques due to their additional force-dipole flow fields. We do not observe clustering for pusher squirmers. Adding a rotlet dipole to the squirmer flow field induces swirling clusters. At high gravitational strengths, the hydrodynamic interactions with the no-slip boundary create an additional vertical alignment for neutral squirmers, which also supports cluster formation. △ Less

Submitted 10 January, 2022; originally announced January 2022.

Comments: contains 9 coloured figures

The scaling potential of beam-splitter-based coherent beam combination

Authors: Michael Mueller, Christopher Aleshire, Joachim Buldt, Henning Stark, Christian Grebing, Arno Klenke, Jens Limpert

Abstract: The impact of nonlinear refraction and residual absorption on the achievable peak and average power in beam-splitter-based coherent beam combination is analyzed theoretically. While the peak power remains limited only by the aperture size, a fundamental average power limit is given by the thermo-optical and thermo-mechanical properties of the beam splitter material and its coatings. Based on our a… ▽ More The impact of nonlinear refraction and residual absorption on the achievable peak and average power in beam-splitter-based coherent beam combination is analyzed theoretically. While the peak power remains limited only by the aperture size, a fundamental average power limit is given by the thermo-optical and thermo-mechanical properties of the beam splitter material and its coatings. Based on our analysis, 100 kW average power can be obtained with state-of-the-art optics at maintained high beam quality (M2 {\leq} 1.1) and at only 2% loss of combining efficiency. This result indicates that the power-scaling potential of today's beam-splitter-based coherent beam combination is far from being depleted. A potential scaling route to megawatt-level average power is discussed for optimized beam splitter geometry. △ Less

Submitted 10 January, 2022; originally announced January 2022.

Journal ref: Opt. Express 29, 27900-27911 (2021)

Multipass cell for high-power few-cycle compression

Authors: Michael Mueller, Joachim Buldt, Henning Stark, Christian Grebing, Jens Limpert

Abstract: A multipass cell for nonlinear compression to few-cycle pulse duration is introduced comprising dielectrically enhanced silver mirrors on silicon substrates. Spectral broadening with 388 W output average power and 776 μJ pulse energy is obtained at 82% cell transmission. A high output beam quality (M2<1.2) and a high spatio-spectral homogeneity (97.5%) as well as the compressibility of the output… ▽ More A multipass cell for nonlinear compression to few-cycle pulse duration is introduced comprising dielectrically enhanced silver mirrors on silicon substrates. Spectral broadening with 388 W output average power and 776 μJ pulse energy is obtained at 82% cell transmission. A high output beam quality (M2<1.2) and a high spatio-spectral homogeneity (97.5%) as well as the compressibility of the output pulses to 6.9 fs duration are demonstrated. Finite element analysis reveals scalability of this cell to 2 kW average output power. △ Less

Submitted 10 January, 2022; originally announced January 2022.

Journal ref: Opt. Lett. 46, 2678-2681 (2021)

Gas-plasma based generation of broadband THz radiation with 640 mW average power

Authors: Joachim Buldt, Henning Stark, Michael Mueller, Christian Grebing, Cesar Jauregui, Jens Limpert

Abstract: We present a high-power source of broadband terahertz radiation covering the whole THz spectral region (0.1-30 THz). The two-color gas plasma generation process is driven by a state-of-the-art Ytterbium fiber chirped pulse amplification system based on coherent combination of 16 rod-type amplifiers. Prior to the THz generation, the pulses are spectrally broadened in a multi-pass-cell and compresse… ▽ More We present a high-power source of broadband terahertz radiation covering the whole THz spectral region (0.1-30 THz). The two-color gas plasma generation process is driven by a state-of-the-art Ytterbium fiber chirped pulse amplification system based on coherent combination of 16 rod-type amplifiers. Prior to the THz generation, the pulses are spectrally broadened in a multi-pass-cell and compressed to 37 fs with a pulse-energy of 1.3 mJ at a repetition rate of 500 kHz. A gas-jet scheme has been employed for the THz generation, increasing the efficiency of the process to 0.1%. The air-biased-coherent-detection scheme is implemented to characterize the full bandwidth of the generated radiation. A THz average power of 640 mW is generated, which is the highest THz average power achieved to date. This makes this source suitable for a variety of applications, e.g. spectroscopy of strongly absorbing samples or driving nonlinear effects for the studies of material properties. △ Less

Submitted 10 January, 2022; originally announced January 2022.

Journal ref: Opt. Lett. 46, 5256-5259 (2021)

3D Infomax improves GNNs for Molecular Property Prediction

Authors: Hannes Stärk, Dominique Beaini, Gabriele Corso, Prudencio Tossou, Christian Dallago, Stephan Günnemann, Pietro Liò

Abstract: Molecular property prediction is one of the fastest-growing applications of deep learning with critical real-world impacts. Including 3D molecular structure as input to learned models improves their performance for many molecular tasks. However, this information is infeasible to compute at the scale required by several real-world applications. We propose pre-training a model to reason about the ge… ▽ More Molecular property prediction is one of the fastest-growing applications of deep learning with critical real-world impacts. Including 3D molecular structure as input to learned models improves their performance for many molecular tasks. However, this information is infeasible to compute at the scale required by several real-world applications. We propose pre-training a model to reason about the geometry of molecules given only their 2D molecular graphs. Using methods from self-supervised learning, we maximize the mutual information between 3D summary vectors and the representations of a Graph Neural Network (GNN) such that they contain latent 3D information. During fine-tuning on molecules with unknown geometry, the GNN still generates implicit 3D information and can use it to improve downstream tasks. We show that 3D pre-training provides significant improvements for a wide range of properties, such as a 22% average MAE reduction on eight quantum mechanical properties. Moreover, the learned representations can be effectively transferred between datasets in different molecular spaces. △ Less

Submitted 4 June, 2022; v1 submitted 8 October, 2021; originally announced October 2021.

Comments: 39th International Conference on Machine Learning (ICML 2022). Also accepted at NeurIPS 2021 ML4PH, AI4S, and SSL workshops and as oral at ELLIS ML4Molecules. 24 pages, 7 figures, 18 tables

Journal ref: 39th International Conference on Machine Learning (ICML 2022)

On the cross-streamline lift of microswimmers in viscoelastic flows

Authors: Akash Choudhary, Holger Stark

Abstract: The current work studies the dynamics of a microswimmer in pressure-driven flow of a weakly viscoelastic fluid. Employing the second-order fluid model, we show that the self-propelling swimmer experiences a viscoelastic swimming lift in addition to the well-known passive lift that arises from its resistance to shear flow. Using the reciprocal theorem, we evaluate analytical expressions for the swi… ▽ More The current work studies the dynamics of a microswimmer in pressure-driven flow of a weakly viscoelastic fluid. Employing the second-order fluid model, we show that the self-propelling swimmer experiences a viscoelastic swimming lift in addition to the well-known passive lift that arises from its resistance to shear flow. Using the reciprocal theorem, we evaluate analytical expressions for the swimming lift experienced by neutral and pusher/puller-type swimmers and show that they depend on the hydrodynamic signature associated with the swimming mechanism. We find that for neutral swimmers focusing towards the centerline is accelerated by two orders of magnitude, while for force-dipole swimmers no net modification in cross-streamline migration occurs. △ Less

Submitted 27 September, 2021; originally announced September 2021.

Oscillatory active microrheology of active suspensions

Authors: M. Knezevic, L. E. Aviles Podgurski, H. Stark

Abstract: Using the method of Brownian dynamics, we investigate the dynamic properties of a 2d suspension of active disks at high Péclet numbers using active microrheology. In our simulations the tracer particle is driven either by a constant or an oscillatory external force. In the first case, we find that the mobility of the tracer initially appreciably decreases with the external force and then becomes a… ▽ More Using the method of Brownian dynamics, we investigate the dynamic properties of a 2d suspension of active disks at high Péclet numbers using active microrheology. In our simulations the tracer particle is driven either by a constant or an oscillatory external force. In the first case, we find that the mobility of the tracer initially appreciably decreases with the external force and then becomes approximately constant for larger forces. For an oscillatory driving force we find that the dynamic mobility shows a quite complex behavior -- it displays a highly nonlinear behavior on both the amplitude and frequency of the driving force. This result is important because it reveals that a phenomenological description of tracer motion in active media in terms of a simple linear stochastic equation even with a memory-mobility kernel is not appropriate. △ Less

Submitted 23 September, 2021; originally announced September 2021.

Comments: 11 pages, 7 figures

How inertial lift affects the dynamics of a microswimmer in Poiseuille flow

Authors: Akash Choudhary, Subhechchha Paul, Felix Rühle, Holger Stark

Abstract: We analyze the dynamics of a microswimmer in pressure-driven Poiseuille flow, where fluid inertia is small but non-negligible. Using perturbation theory and the reciprocal theorem, we show that in addition to the classical inertial lift of passive particles, the active nature generates a `swimming lift', which we evaluate for neutral and pusher/puller-type swimmers. Accounting for fluid inertia en… ▽ More We analyze the dynamics of a microswimmer in pressure-driven Poiseuille flow, where fluid inertia is small but non-negligible. Using perturbation theory and the reciprocal theorem, we show that in addition to the classical inertial lift of passive particles, the active nature generates a `swimming lift', which we evaluate for neutral and pusher/puller-type swimmers. Accounting for fluid inertia engenders a rich spectrum of novel complex dynamics including bistable states, where tumbling coexists with stable centerline swimming or swinging. The dynamics is sensitive to the swimmer's hydrodynamic signature and goes well beyond the findings at vanishing fluid inertia. Our work will have non-trivial implications on the transport and dispersion of active suspensions in microchannels. △ Less

Submitted 9 September, 2021; originally announced September 2021.

Jointly Learnable Data Augmentations for Self-Supervised GNNs

Authors: Zekarias T. Kefato, Sarunas Girdzijauskas, Hannes Stärk

Abstract: Self-supervised Learning (SSL) aims at learning representations of objects without relying on manual labeling. Recently, a number of SSL methods for graph representation learning have achieved performance comparable to SOTA semi-supervised GNNs. A Siamese network, which relies on data augmentation, is the popular architecture used in these methods. However, these methods rely on heuristically craf… ▽ More Self-supervised Learning (SSL) aims at learning representations of objects without relying on manual labeling. Recently, a number of SSL methods for graph representation learning have achieved performance comparable to SOTA semi-supervised GNNs. A Siamese network, which relies on data augmentation, is the popular architecture used in these methods. However, these methods rely on heuristically crafted data augmentation techniques. Furthermore, they use either contrastive terms or other tricks (e.g., asymmetry) to avoid trivial solutions that can occur in Siamese networks. In this study, we propose, GraphSurgeon, a novel SSL method for GNNs with the following features. First, instead of heuristics we propose a learnable data augmentation method that is jointly learned with the embeddings by leveraging the inherent signal encoded in the graph. In addition, we take advantage of the flexibility of the learnable data augmentation and introduce a new strategy that augments in the embedding space, called post augmentation. This strategy has a significantly lower memory overhead and run-time cost. Second, as it is difficult to sample truly contrastive terms, we avoid explicit negative sampling. Third, instead of relying on engineering tricks, we use a scalable constrained optimization objective motivated by Laplacian Eigenmaps to avoid trivial solutions. To validate the practical use of GraphSurgeon, we perform empirical evaluation using 14 public datasets across a number of domains and ranging from small to large scale graphs with hundreds of millions of edges. Our finding shows that GraphSurgeon is comparable to six SOTA semi-supervised and on par with five SOTA self-supervised baselines in node classification tasks. The source code is available at https://github.com/zekarias-tilahun/graph-surgeon. △ Less

Submitted 23 August, 2021; originally announced August 2021.

Comments: Under Review

Perspectives on viscoelastic flow instabilities and elastic turbulence

Authors: Sujit S. Datta, Arezoo M. Ardekani, Paulo E. Arratia, Antony N. Beris, Irmgard Bischofberger, Jens G. Eggers, J. Esteban López-Aguilar, Suzanne M. Fielding, Anna Frishman, Michael D. Graham, Jeffrey S. Guasto, Simon J. Haward, Sarah Hormozi, Gareth H. McKinley, Robert J. Poole, Alexander Morozov, V. Shankar, Eric S. G. Shaqfeh, Amy Q. Shen, Holger Stark, Victor Steinberg, Ganesh Subramanian, Howard A. Stone

Abstract: Viscoelastic fluids are a common subclass of rheologically complex materials that are encountered in diverse fields from biology to polymer processing. Often the flows of viscoelastic fluids are unstable in situations where ordinary Newtonian fluids are stable, owing to the nonlinear coupling of the elastic and viscous stresses. Perhaps more surprisingly, the instabilities produce flows with the h… ▽ More Viscoelastic fluids are a common subclass of rheologically complex materials that are encountered in diverse fields from biology to polymer processing. Often the flows of viscoelastic fluids are unstable in situations where ordinary Newtonian fluids are stable, owing to the nonlinear coupling of the elastic and viscous stresses. Perhaps more surprisingly, the instabilities produce flows with the hallmarks of turbulence -- even though the effective Reynolds numbers may be $O(1)$ or smaller. We provide perspectives on viscoelastic flow instabilities by integrating the input from speakers at a recent international workshop: historical remarks, characterization of fluids and flows, discussion of experimental and simulation tools, and modern questions and puzzles that motivate further studies of this fascinating subject. The materials here will be useful for researchers and educators alike, especially as the subject continues to evolve in both fundamental understanding and applications in engineering and the sciences. △ Less

Submitted 22 August, 2021; originally announced August 2021.

Comments: A perspective article based on a virtual workshop of the Princeton Center for Theoretical Sciences that took place in January 2021

Droplets on substrates with oscillating wettability

Authors: Josua Grawitter, Holger Stark

Abstract: In recent decades novel solid substrates have been designed which change their wettability in response to light or an electrostatic field. Here, we investigate a droplet on substrates with oscillating uniform wettability by varying minimium and maximum contact angles and frequency. To simulate this situation, we use our previous work [Grawitter and Stark, Soft Matter 17, 2454 (2021)], where we imp… ▽ More In recent decades novel solid substrates have been designed which change their wettability in response to light or an electrostatic field. Here, we investigate a droplet on substrates with oscillating uniform wettability by varying minimium and maximum contact angles and frequency. To simulate this situation, we use our previous work [Grawitter and Stark, Soft Matter 17, 2454 (2021)], where we implemented the boundary element method in combination with the Cox-Voinov law for the contact-line velocity, to determine the fluid flow inside a droplet. After a transient regime the droplet performs steady oscillations, the amplitude of which decreases with increasing frequency. For slow oscillations our numerical results agree well with the linearized spherical-cap model. They collapse on a master curve when we rescale frequency by a characteristic relaxation time. In contrast, for fast oscillations we observe significant deviations from the master curve. The decay of the susceptibility is weaker and the phase shift between oscillations in wettability and contact angle stays below the predicted $π/2$. The reason becomes obvious when studying the combined dynamics of droplet height and contact angle. It reveals non-reciprocal shape changes during one oscillation period even at low frequencies due to the induced fluid flow inside the droplet, which are not captured by the spherical-cap model. Similar periodic non-reciprocal shape changes occur at low frequencies when the droplet is placed on an oscillating nonuniform wettability profile with six-fold symmetry. Such profiles are inspired by the light intensity pattern of Laguerre-Gauss laser modes. Since the non-reciprocal shape changes induce fluid circulation, which is controllable from the outside, our findings envisage to design targeted microfluidic transport of solutes inside the droplet. △ Less

Submitted 3 August, 2021; originally announced August 2021.

Comments: 12 pages, 10 figures

Journal ref: Soft Matter 17 (2021) 9469-9479

Turbulence-induced clustering in compressible active fluids

Authors: Vasco M. Worlitzer, Gil Ariel, Avraham Be'er, Holger Stark, Markus Bär, Sebastian Heidenreich

Abstract: We study a novel phase of active polar fluids, which is characterized by the continuous creation and destruction of dense clusters due to self-sustained turbulence. This state arises due to the interplay of the self-advection of the aligned swimmers and their defect topology. The typical cluster size is determined by the characteristic vortex size. Our results are obtained by investigating a conti… ▽ More We study a novel phase of active polar fluids, which is characterized by the continuous creation and destruction of dense clusters due to self-sustained turbulence. This state arises due to the interplay of the self-advection of the aligned swimmers and their defect topology. The typical cluster size is determined by the characteristic vortex size. Our results are obtained by investigating a continuum model of compressible polar active fluids, which incorporates typical experimental observations in bacterial suspensions by assuming a non-monotone dependence of speed on density. △ Less

Submitted 15 July, 2021; originally announced July 2021.

Comments: supplementary information available on demand

Journal ref: Soft Matter 2021