Enhanced force-field calibration via machine learning

The influence of microscopic force fields on the motion of Brownian particles plays a fundamental role in a broad range of fields, including soft matter, biophysics, and active matter. Often, the experimental calibration of these force fields relies on the analysis of the trajectories of these Brownian particles. However, such an analysis is not always straightforward, especially if the underlying force fields are non-conservative or time-varying, driving the system out of thermodynamic equilibrium. Here, we introduce a toolbox to calibrate microscopic force fields by analyzing the trajectories of a Brownian particle using machine learning, namely recurrent neural networks. We demonstrate that this machine-learning approach outperforms standard methods when characterizing the force fields generated by harmonic potentials if the available data are limited. More importantly, it provides a tool to calibrate force fields in situations for which there are no standard methods, such as non-conservative and time-varying force fields. In order to make this method readily available for other users, we provide a Python software package named DeepCalib, which can be easily personalized and optimized for specific applications.

[1]  Quoc V. Le,et al.  Don't Decay the Learning Rate, Increase the Batch Size , 2017, ICLR.

[2]  Shoichi Toyabe,et al.  Nonequilibrium energetics of a single F1-ATPase molecule. , 2010, Physical review letters.

[3]  Isaac C. D. Lenton,et al.  Machine learning reveals complex behaviours in optically trapped particles , 2020, Mach. Learn. Sci. Technol..

[4]  Giorgio Volpe,et al.  High-performance reconstruction of microscopic force fields from Brownian trajectories , 2018, Nature Communications.

[5]  John Bechhoefer,et al.  Real-time calibration of a feedback trap. , 2014, The Review of scientific instruments.

[6]  François Chollet,et al.  Keras: The Python Deep Learning library , 2018 .

[7]  Frank Cichos,et al.  Thermo-Osmotic Flow in Thin Films. , 2016, Physical review letters.

[8]  Clemens Bechinger,et al.  Realization of a micrometre-sized stochastic heat engine , 2011, Nature Physics.

[9]  D. Petrov,et al.  Brownian Carnot engine , 2014, Nature Physics.

[10]  Muhammad Sahimi,et al.  Approaching complexity by stochastic methods: From biological systems to turbulence , 2011 .

[11]  C. Jarzynski,et al.  Verification of the Crooks fluctuation theorem and recovery of RNA folding free energies , 2005, Nature.

[12]  Frank Cichos,et al.  Optically controlled thermophoretic trapping of single nano-objects. , 2013, ACS nano.

[13]  Frank Cichos,et al.  Single Molecules Trapped by Dynamic Inhomogeneous Temperature Fields. , 2015, Nano letters.

[14]  Maciej Lewenstein,et al.  Single trajectory characterization via machine learning , 2020, New Journal of Physics.

[15]  R. Simmons,et al.  Elasticity of the red cell membrane and its relation to hemolytic disorders: an optical tweezers study. , 1999, Biophysical journal.

[16]  David R. Smith,et al.  Interparticle Coupling Effects on Plasmon Resonances of Nanogold Particles , 2003 .

[17]  G. Kane.  Parallel Distributed Processing: Explorations in the Microstructure of Cognition, vol 1: Foundations, vol 2: Psychological and Biological Models , 1994 .

[18]  G. Volpe,et al.  Simulation of a Brownian particle in an optical trap , 2013 .

[19]  Yoav Shechtman,et al.  Single-Particle Diffusion Characterization by Deep Learning. , 2019, Biophysical journal.

[20]  Giovanni Volpe,et al.  Digital video microscopy enhanced by deep learning , 2018, Optica.

[21]  Frank Cichos,et al.  Machine learning for active matter , 2020, Nat. Mach. Intell..

[22]  Giorgio Volpe,et al.  Brownian motion in a nonhomogeneous force field and photonic force microscope. , 2007, Physical review. E, Statistical, nonlinear, and soft matter physics.

[23]  Kishan Dholakia,et al.  Optoelectronic tweezers , 2005, Nature materials.

[24]  Song Han,et al.  ESE: Efficient Speech Recognition Engine with Sparse LSTM on FPGA , 2016, FPGA.

[25]  Lenka Zdeborová,et al.  New tool in the box , 2017, Nature Physics.

[26]  Gerhard Hummer,et al.  Free energy profiles from single-molecule pulling experiments , 2010, Proceedings of the National Academy of Sciences.

[27]  Martin Fränzl,et al.  Thermophoretic trap for single amyloid fibril and protein aggregation studies , 2019, Nature Methods.

[28]  Alireza Seif,et al.  Machine learning the thermodynamic arrow of time , 2019, Nature Physics.

[29]  S. Suresh,et al.  Nonlinear elastic and viscoelastic deformation of the human red blood cell with optical tweezers. , 2004, Mechanics & chemistry of biosystems : MCB.

[30]  Alois Würger,et al.  Thermal non-equilibrium transport in colloids , 2010 .

[31]  Yonggun Jun,et al.  High-precision test of Landauer's principle in a feedback trap. , 2014, Physical review letters.

[32]  H. Flyvbjerg,et al.  Power spectrum analysis for optical tweezers , 2004 .

[33]  Hiroshi Yokoyama,et al.  Direct observation of anisotropic interparticle forces in nematic colloids with optical tweezers. , 2004, Physical review letters.

[34]  E. Lutz,et al.  Experimental verification of Landauer’s principle linking information and thermodynamics , 2012, Nature.

[35]  Jean-Baptiste Masson,et al.  A Bayesian inference scheme to extract diffusivity and potential fields from confined single-molecule trajectories. , 2012, Biophysical journal.

[36]  Yoav Shechtman,et al.  Single particle diffusion characterization by deep learning , 2019, bioRxiv.

[37]  Ralf Eichhorn,et al.  Experimental realization of a minimal microscopic heat engine. , 2017, Physical review. E.

[38]  Mark D Hannel,et al.  Machine-learning techniques for fast and accurate feature localization in holograms of colloidal particles. , 2018, Optics express.

[39]  Alejandro V. Arzola,et al.  Optical Tweezers: A Comprehensive Tutorial from Calibration to Applications , 2020 .

[40]  T Speck,et al.  Einstein relation generalized to nonequilibrium. , 2007, Physical review letters.

[41]  I. Tinoco,et al.  Equilibrium Information from Nonequilibrium Measurements in an Experimental Test of Jarzynski's Equality , 2002, Science.

[42]  Joachim Peinke,et al.  Reconstruction of complex dynamical systems affected by strong measurement noise. , 2006, Physical review letters.

[43]  Juan Ruben Gomez-Solano,et al.  Experimental verification of a modified fluctuation-dissipation relation for a micron-sized particle in a nonequilibrium steady state. , 2009, Physical review letters.

[44]  D. Herschlag,et al.  Direct Measurement of the Full, Sequence-Dependent Folding Landscape of a Nucleic Acid , 2006, Science.

[45]  Navdeep Jaitly,et al.  Hybrid speech recognition with Deep Bidirectional LSTM , 2013, 2013 IEEE Workshop on Automatic Speech Recognition and Understanding.

[46]  Mark Dykman,et al.  Thermally activated transitions in a bistable three-dimensional optical trap , 1999, Nature.

[47]  Zachary Chase Lipton.  A Critical Review of Recurrent Neural Networks for Sequence Learning , 2015, ArXiv.

[48]  Pierre Ronceray,et al.  Learning Force Fields from Stochastic Trajectories , 2018, Physical Review X.

[49]  Jürgen Schmidhuber,et al.  Learning to Forget: Continual Prediction with LSTM , 2000, Neural Computation.

[50]  Édgar Roldán,et al.  Colloidal heat engines: a review. , 2017, Soft Matter.

[51]  Ralf Eichhorn,et al.  Measurement of anomalous diffusion using recurrent neural networks. , 2019, Physical review. E.

[52]  Alexandr Jonas,et al.  Direct measurement of the nonconservative force field generated by optical tweezers. , 2009, Physical review letters.

[53]  Frank Cichos,et al.  Microscopic engine powered by critical demixing , 2017 .

[54]  James L. McClelland,et al.  Parallel distributed processing: explorations in the microstructure of cognition, vol. 1: foundations , 1986 .

[55]  Giovanni Volpe,et al.  Effective drifts in dynamical systems with multiplicative noise: a review of recent progress , 2016, Reports on progress in physics. Physical Society.

[56]  G. Volpe,et al.  Nonadditivity of critical Casimir forces , 2015, Nature Communications.

[57]  G. Volpe,et al.  Torque detection using Brownian fluctuations. , 2006, Physical review letters.

[58]  Halina Rubinsztein-Dunlop,et al.  Machine learning wall effects of eccentric spheres for convenient computation. , 2019, Physical review. E.

[59]  P. Quinto-Su,et al.  A microscopic steam engine implemented in an optical tweezer , 2014, Nature Communications.

[60]  A. Ozcan,et al.  On the use of deep learning for computational imaging , 2019, Optica.

[61]  S. Ciliberto,et al.  Experiments in Stochastic Thermodynamics: Short History and Perspectives , 2017 .

[62]  George Kurian,et al.  Google's Neural Machine Translation System: Bridging the Gap between Human and Machine Translation , 2016, ArXiv.

[63]  Ayan Banerjee,et al.  Fast Bayesian inference of optical trap stiffness and particle diffusion , 2016, Scientific Reports.

[64]  Haw Yang,et al.  Quantitative characterization of changes in dynamical behavior for single-particle tracking studies. , 2006, Journal of Physical Chemistry B.