Visit our GitHub page at https://github.com/bozdaglab/ for the full list.
SUPREME (a subtype prediction methodology) integrates multi-omics datasets in a similarity network and learn node embeddings to perform subtype classification of patients.
PhenoGeneRanker is a gene/disease prioritization tool that utilizes heterogeneous gene disease network. PhenoGeneRanker allows multi-layer gene and disease networks. It also calculates empirical p-values of gene ranking using random stratified sampling of genes based on their connectivity degree in the network.
miRDriver is a computational pipeline that integrates gene and miRNA expression, copy number alteration, DNA methylation and TF-gene interaction information to infer copy number-derived miRNA-gene networks in cancer.
A computational pipeline to infer competing endogenous RNA (ceRNA) interactions in cancer.
ProcessDriver is a tool to detect copy number-based drivers and associated biological processes in cancer.
A canonical correlation analysis based algorithm that utilizes DNA methylation, copy number alteration, gene expression datasets from cancer samples to infer regulatory interactions between genes.
A feature selection algorithm to find most predictive methylation probes for gene expression.
R/GSEPD is a bioinformatics package for R to help disambiguate transcriptome samples (a matrix of RNA-Seq counts at RefSeq IDs) by automating differential expression (with DESeq2), then gene set enrichment (with GOSeq), and finally a N-dimensional projection to quantify in which ways each sample is like either treatment group.
FastMEDUSA is a parallel program to infer gene regulatory networks from gene expression and promoter sequences. It is the parallelized version of MEDUSA (Kundaje, et al. 2008 Plos Comp. Bio.)
FMTP is a software tool to compute the MTP of a physical map based purely on restriction fingerprint data (and the contigs). FMTP completely ignores the ordering of clones obtained by the physical map algorithm.
Compartmentalized assembler is a novel method for the assemlby of high quality physical maps from fingerprinted clones. Our method exploits the presence of genetic markers at the genomic level that allows us to pre-cluster the clones. For each cluster of clones, a local physical map is first constructed. Then, all the individual maps are carefully merged into the final physical map.