scholarly journals Covariant Fitness Clusters Reveal Structural Evolution of SARS-CoV-2 Polymerase Across the Human Population

2022 ◽  
Author(s):  
Chao Wang ◽  
Nadia Elghobashi-Meinhardt ◽  
William E Balch
Keyword(s):  
Fitness Landscape ◽  
Structural Evolution ◽  
Gaussian Process Regression ◽  
Binding Motif ◽  
Learning Approach ◽  
Spatial Covariance ◽  
Evolutionary Mechanisms ◽  
Viral Genomes ◽  
Machine Learning Approach ◽  
Rapid Spread

Understanding the fitness landscape of viral mutations is crucial for uncovering the evolutionary mechanisms contributing to pandemic behavior. Here, we apply a Gaussian process regression (GPR) based machine learning approach that generates spatial covariance (SCV) relationships to construct stability fitness landscapes for the RNA-dependent RNA polymerase (RdRp) of SARS-CoV-2. GPR generated fitness scores capture on a residue-by-residue basis a covariant fitness cluster centered at the C487-H642-C645-C646 Zn2+ binding motif that iteratively evolves since the early phase pandemic. In the Alpha and Delta variant of concern (VOC), multi-residue SCV interactions in the NiRAN domain form a second fitness cluster contributing to spread. Strikingly, a novel third fitness cluster harboring a Delta VOC basal mutation G671S augments RdRp structural plasticity to potentially promote rapid spread through viral load. GPR principled SCV provides a generalizable tool to mechanistically understand evolution of viral genomes at atomic resolution contributing to fitness at the pathogen-host interface.

Machine learning approach for accurate backmapping of coarse-grained models to all-atom models

2020 ◽  
Vol 56 (65) ◽  
pp. 9312-9315 ◽  
Author(s):  
Yaxin An ◽  
Sanket A. Deshmukh
Keyword(s):  
Neural Network ◽  
Machine Learning ◽  
Artificial Neural Network ◽  
Random Forest ◽  
Regression Models ◽  
Gaussian Process Regression ◽  
Nearest Neighbors ◽  
Coarse Grained ◽  
Learning Approach ◽  
Machine Learning Approach

Four different machine learning (ML) regression models: artificial neural network, k-nearest neighbors, Gaussian process regression and random forest were built to backmap coarse-grained models to all-atom models.

1552-P: Machine Learning Approach to Decision-Making for Initial Insulin Use in Japanese Patients with Type 2 Diabetes

Diabetes ◽  
2020 ◽  
Vol 69 (Supplement 1) ◽  
pp. 1552-P
Author(s):  
KAZUYA FUJIHARA ◽  
MAYUKO H. YAMADA ◽  
YASUHIRO MATSUBAYASHI ◽  
MASAHIKO YAMAMOTO ◽  
TOSHIHIRO IIZUKA ◽  
...  
Keyword(s):  
Machine Learning ◽  
Type 2 Diabetes ◽  
Decision Making ◽  
Japanese Patients ◽  
Learning Approach ◽  
Machine Learning Approach ◽  
Insulin Use

A Machine Learning Approach to Jet-Surface Interaction Noise Modeling

2020 ◽  
Author(s):  
Clifford A. Brown ◽  
Jonny Dowdall ◽  
Brian Whiteaker ◽  
Lauren McIntyre
Keyword(s):  
Machine Learning ◽  
Surface Interaction ◽  
Learning Approach ◽  
Noise Modeling ◽  
Machine Learning Approach

Mol2vec: Unsupervised Machine Learning Approach with Chemical Intuition

2017 ◽  
Author(s):  
Sabrina Jaeger ◽  
Simone Fulle ◽  
Samo Turk
Keyword(s):  
Machine Learning ◽  
Language Processing ◽  
Supervised Machine Learning ◽  
Learning Approach ◽  
Learning Approaches ◽  
Unsupervised Machine Learning ◽  
Feature Representations ◽  
Machine Learning Approach ◽  
The Individual ◽  
Vector Representations

Inspired by natural language processing techniques we here introduce Mol2vec which is an unsupervised machine learning approach to learn vector representations of molecular substructures. Similarly, to the Word2vec models where vectors of closely related words are in close proximity in the vector space, Mol2vec learns vector representations of molecular substructures that are pointing in similar directions for chemically related substructures. Compounds can finally be encoded as vectors by summing up vectors of the individual substructures and, for instance, feed into supervised machine learning approaches to predict compound properties. The underlying substructure vector embeddings are obtained by training an unsupervised machine learning approach on a so-called corpus of compounds that consists of all available chemical matter. The resulting Mol2vec model is pre-trained once, yields dense vector representations and overcomes drawbacks of common compound feature representations such as sparseness and bit collisions. The prediction capabilities are demonstrated on several compound property and bioactivity data sets and compared with results obtained for Morgan fingerprints as reference compound representation. Mol2vec can be easily combined with ProtVec, which employs the same Word2vec concept on protein sequences, resulting in a proteochemometric approach that is alignment independent and can be thus also easily used for proteins with low sequence similarities.

Sign in / Sign up

Export Citation Format

Share Document