PMID- 34213915
OWN - NLM
STAT- PubMed-not-MEDLINE
DCOM- 20210727
LR  - 20240403
IS  - 1520-5215 (Electronic)
IS  - 1089-5639 (Print)
IS  - 1089-5639 (Linking)
VI  - 125
IP  - 28
DP  - 2021 Jul 22
TI  - Multiscale Reweighted Stochastic Embedding: Deep Learning of Collective Variables 
      for Enhanced Sampling.
PG  - 6286-6302
LID - 10.1021/acs.jpca.1c02869 [doi]
AB  - Machine learning methods provide a general framework for automatically finding 
      and representing the essential characteristics of simulation data. This task is 
      particularly crucial in enhanced sampling simulations. There we seek a few 
      generalized degrees of freedom, referred to as collective variables (CVs), to 
      represent and drive the sampling of the free energy landscape. In theory, these 
      CVs should separate different metastable states and correspond to the slow 
      degrees of freedom of the studied physical process. To this aim, we propose a new 
      method that we call multiscale reweighted stochastic embedding (MRSE). Our work 
      builds upon a parametric version of stochastic neighbor embedding. The technique 
      automatically learns CVs that map a high-dimensional feature space to a 
      low-dimensional latent space via a deep neural network. We introduce several new 
      advancements to stochastic neighbor embedding methods that make MRSE especially 
      suitable for enhanced sampling simulations: (1) weight-tempered random sampling 
      as a landmark selection scheme to obtain training data sets that strike a balance 
      between equilibrium representation and capturing important metastable states 
      lying higher in free energy; (2) a multiscale representation of the 
      high-dimensional feature space via a Gaussian mixture probability model; and (3) 
      a reweighting procedure to account for training data from a biased probability 
      distribution. We show that MRSE constructs low-dimensional CVs that can correctly 
      characterize the different metastable states in three model systems: the 
      Muller-Brown potential, alanine dipeptide, and alanine tetrapeptide.
FAU - Rydzewski, Jakub
AU  - Rydzewski J
AUID- ORCID: 0000-0003-4325-4177
AD  - Institute of Physics, Faculty of Physics, Astronomy and Informatics, Nicolaus 
      Copernicus University, Grudziadzka 5, 87-100 Torun, Poland.
FAU - Valsson, Omar
AU  - Valsson O
AUID- ORCID: 0000-0001-7971-4767
AD  - Max Planck Institute for Polymer Research, Ackermannweg 10, Mainz D-55128, 
      Germany.
LA  - eng
PT  - Journal Article
DEP - 20210702
PL  - United States
TA  - J Phys Chem A
JT  - The journal of physical chemistry. A
JID - 9890903
SB  - IM
PMC - PMC8389995
COIS- The authors declare no competing financial interest.
EDAT- 2021/07/03 06:00
MHDA- 2021/07/03 06:01
PMCR- 2021/08/26
CRDT- 2021/07/02 17:11
PHST- 2021/07/03 06:00 [pubmed]
PHST- 2021/07/03 06:01 [medline]
PHST- 2021/07/02 17:11 [entrez]
PHST- 2021/08/26 00:00 [pmc-release]
AID - 10.1021/acs.jpca.1c02869 [doi]
PST - ppublish
SO  - J Phys Chem A. 2021 Jul 22;125(28):6286-6302. doi: 10.1021/acs.jpca.1c02869. Epub 
      2021 Jul 2.