scGNN is an iterative approach using three multi-modal autoencoders for gene imputation and cell clustering using graph neural networks integrated with LTMG model of gene expression in single cell RNA-seq analysis.
Based on a graph neural network, NRI-MD is a neural relational inference model adopts an encoder-decoder architecture to simultaneously infer latent interactions for probing protein allosteric processes as dynamic networks of interacting residues. From the MD trajectories, this model successfully learned the long-range interactions and pathways that can mediate the allosteric communications. Furthermore, the model can discover allostery-related interactions earlier in the MD simulation trajectories and predict relative free energy changes upon mutations more accurately than other methods..
GRGNN is an end-to-end approach to reconstruct gene regulatory networks from scratch utilizing the gene expression data, in both a supervised and a semi-supervised framework..
An auto-updating web-service to annotate genomic variants on protein structures. Since 2016, G2S is official API supports structure mapping services in cBioPortal for cancer genomics.
Bayesian High-order Interaction Toolkit for genotype-phenotype epistasis inference.