REVERT : REVERse Transition
REVERT : REVERse Transition
Embedded Files
The REVERT model leverages state-of-the-art attractor landscape analysis to identify molecular reversion switches in colorectal cancer. By integrating Boolean network modeling with single-cell transcriptomic data, REVERT reconstructs dynamic gene regulatory networks (GRNs) and predicts optimal intervention targets for reversing tumorigenesis. Through in silico perturbation analysis, REVERT identifies key transcription factors, including YY1 and MYC, as major regulators and highlights USP7 as a promising therapeutic target. Experimental validation using colorectal cancer organoids confirms its efficacy in shifting cancer cells toward normal-like states. This tool represents a major advancement in cancer reversion research, providing a powerful and efficient in silico solution for identifying reprogramming targets in tumorigenesis.
For more details, refer to the publication and github:
Con_ACP : Contrastive Learning-Based Model for Anticancer Peptides
Con_ACP : Contrastive Learning-Based Model for Anticancer Peptides
The Con_ACP model leverages state-of-the-art contrastive learning techniques to enhance the extraction of features from peptide sequences, enabling accurate prediction of anticancer peptides (ACPs). By utilizing dual-tokenizer encoding and innovative loss functions, Con_ACP achieves superior performance across benchmark datasets compared to previous models. This tool represents a significant step forward in peptide-based therapeutic discovery, offering a cost-effective and efficient in silico solution for identifying promising ACP candidates.
For more details, refer to the publication and github:
NetG2P : Network-based genotype-to-phenotype transformation
NetG2P : Network-based genotype-to-phenotype transformation
NetG2P is an innovative computational framework that leverages network propagation and explainable AI to translate genomic data into actionable pathway interaction networks. By identifying critical oncogenic features (COFs) and signaling crosstalk, NetG2P provides insights into cancer prognosis and potential therapeutic targets. This model has been applied across ten cancer types using The Cancer Genome Atlas (TCGA) dataset, enabling personalized treatment strategies and advancing precision medicine in oncology.
For more details, refer to the publication and github:
scVarID : Mapping Genetic Variants to Single Cells for Cancer Research
scVarID : Mapping Genetic Variants to Single Cells for Cancer Research
scVarID is a cutting-edge computational tool designed to map genetic variants to single cells, leveraging the power of single-cell long-read RNA sequencing (scLR-seq). By integrating variant allele frequencies (VAFs) with gene expression data, scVarID provides unprecedented insights into clonal heterogeneity and allele-specific expression in cancer and other genetic diseases.
For more details, refer to the publication and github:
Page updated
Google Sites
Report abuse