2020
DOI: 10.1016/j.bpj.2019.12.016
|View full text |Cite
|
Sign up to set email alerts
|

Molecular Insights from Conformational Ensembles via Machine Learning

Abstract: Biomolecular simulations are intrinsically high dimensional and generate noisy data sets of ever-increasing size. Extracting important features from the data is crucial for understanding the biophysical properties of molecular processes, but remains a big challenge. Machine learning (ML) provides powerful dimensionality reduction tools. However, such methods are often criticized as resembling black boxes with limited human-interpretable insight. We use methods from supervised and unsupervised ML to efficiently… Show more

Help me understand this report
View preprint versions

Search citation statements

Order By: Relevance

Paper Sections

Select...
2
1
1
1

Citation Types

0
100
0

Year Published

2020
2020
2024
2024

Publication Types

Select...
4
3
2

Relationship

0
9

Authors

Journals

citations
Cited by 82 publications
(100 citation statements)
references
References 78 publications
0
100
0
Order By: Relevance
“…Association of sclx 8 to Cytc has a local impact on the flexibility of the residues involving in the binding site but, because of the tertiary structure of protein, this effect can be propagated on a large distance. To identify the possible allosteric effects upon sclx 8 we took advantage of a supervised machine-learning protocol recently proposed by Fleetwood and coworkers 61 to monitor the structural changes of the cytochrome upon binding of sclx 8 . Important residues known to interact with sclx 8 correspond to sites 1, 2 and 3 are denoted in Figure 5 with stars and the respective site color (green, red and gray).…”
Section: Resultsmentioning
confidence: 99%
See 1 more Smart Citation
“…Association of sclx 8 to Cytc has a local impact on the flexibility of the residues involving in the binding site but, because of the tertiary structure of protein, this effect can be propagated on a large distance. To identify the possible allosteric effects upon sclx 8 we took advantage of a supervised machine-learning protocol recently proposed by Fleetwood and coworkers 61 to monitor the structural changes of the cytochrome upon binding of sclx 8 . Important residues known to interact with sclx 8 correspond to sites 1, 2 and 3 are denoted in Figure 5 with stars and the respective site color (green, red and gray).…”
Section: Resultsmentioning
confidence: 99%
“…Recently, ML methods have been applied to learn ensemble properties from molecular simulations and to provide easily interpretable metrics of important features. In this study, we have performed an analysis of our trajectories with Multilayer Perceptrons (MLP) by utilizing the demystifying package from Fleetwood et al 61 . The MLP is a fully connected artificial neural network (ANN) with one input layer, one output layer and at least one hidden layer.…”
Section: Multilayer Perceptronsmentioning
confidence: 99%
“…The magnitudes of the eigenvalues indicates the data variance in each of these directions of motions, the eigenvector with the largest eigenvalue is called the first principal component, and so on. The PCA was performed with a home brew script utilizing the Scikit-learn 57 library, the internal coordinates (inverse distance between geometric centers of two residues) of the trajectory as input instead of Cartesian coordinates, due to better performance 58 . To obtain the per residue importance, the sum of the weighted principal components up to certain threshold with the corresponding as weights is taken.…”
Section: Methodsmentioning
confidence: 99%
“…The magnitudes of the eigenvalues indicates the data variance in each of these directions of motions, the eigenvector with the largest eigenvalue is called the first principal component, and so on. The PCA was performed with a home brew script utilizing the Scikit-learn (36) library, the internal coordinates (inverse distance between geometric centers of two residues) of the trajectory as input instead of Cartesian coordinates, due to better performance (37). To obtain the per residue importance, the sum of the weighted principal components up to certain threshold with the corresponding as weights is taken.…”
Section: Methodsmentioning
confidence: 99%