Loading Events

« All Events

  • This event has passed.

SS 2018: Machine learning for molecular world

17. April 2018
3:00 pm - 4:30 pm


Berliner Str. 41-49
Heidelberg, 69120

by Frauke Gräter and Rebecca Wade

Over the last decade machine learning revolutionized computer vision and language processing.
This progress was fueled by the development of new methods as well as the availability of ever powerful hardware.
Now a new wave of research adapts these advances to increase our understanding of the molecular world.
In this seminar we will explore the recent literature on these efforts ranging from protein dynamics and drug design to materials science.
The seminar is targeted towards advanced Bachelor, Master and interested PhD students. As the seminar’s topics cover a broad range, we are happy to welcome students from all scientific backgrounds with a strong interest in interdisciplinary work, preferably with some background in machine learning, molecular simulations, or quantum chemistry.

Time: Tue, 3-4.30pm,
Place: Mathematikon, INF 205, SR11

In our Vorbesprechung on Tue, April 17, 2018, 3-4.30pm, we will distribute topics and dates. The first ~2 seminars will be used to introduce some basics of machine learning and relevant molecular simulation techniques.


As we only can accept a limited number of participants, please send us an email to frauke.graeter@h-its.org before the course starts, if you want to make sure to be accepted.

Credit points:

– master physics: 6 CP

– master molecular biotechnology and other masters in biology: 4 CP

Recommended reading:

Andrew R. Leach, “Molecular Modeling: Principles and Applications”
Gareth James, Daniela Witten, Trevor Hastie and Robert Tibshirani, “An Introduction to Statistical Learning” http://www-bcf.usc.edu/~gareth/ISL/
Ian Goodfellow and Yoshua Bengio and Aaron Courville, “Deep Learning Book” http://www.deeplearningbook.org/

Preliminary list of references:

Machine Learning Force Field Parameters from Ab Initio Data.
Li Y, Li H, Pickard FC 4th, Narayanan B, Sen FG, Chan MKY, Sankaranarayanan SKRS, Brooks BR, Roux B.
J Chem Theory Comput. 2017 Sep 12;13(9):4492-4503. doi: 10.1021/acs.jctc.7b00521.

Neural network based prediction of conformational free energies – a new route towards coarse-grained simulation models
T. Lemke and C. Peter,
J. Chem. Theory Comput., Just Accepted Manuscript, 2017.

VAMPnets: Deep learning of molecular kinetics.
Nat. Comm., 9 . p. 5.
Mardt, A. and Pasquali, L. and Wu, H. and Noé, F. 2018

The face of crystals: insightful classification using deep learning
A Ziletti, D Kumar, M Scheffler, LM Ghiringhelli
arXiv preprint arXiv:1709.02298 2017

Machine learning of accurate energy-conserving molecular force fields
S Chmiela, A Tkatchenko, HE Sauceda, I Poltavsky, KT Schütt, KR Müller
Science advances 3 (5), e1603015 2017

Quantum-chemical insights from deep tensor neural networks
KT Schütt, F Arbabzadah, S Chmiela, KR Müller, A Tkatchenko
Nature communications 8, 13890 2017

Machine Learning Energies of 2 Million Elpasolite (A B C 2 D 6) Crystals
FA Faber, A Lindmaa, OA Von Lilienfeld, R Armiento
Physical review letters 117 (13), 135502, 2016

Big data meets quantum chemistry approximations: the Δ-machine learning approach
R Ramakrishnan, PO Dral, M Rupp, OA von Lilienfeld
Journal of chemical theory and computation 11 (5), 2087-2096, 2015

Molecular Dynamics Simulations with Quantum Mechanics/Molecular Mechanics and Adaptive Neural Networks
Lin Shen and Weitao Yang
J. Chem. Theory Comput., Article ASAP, Feb. 13, 2018

Click here to go to the German home page.