ISDTool: a computational model for predicting immunosuppressive domain of HERVs

Area of Science:

• * Molecular biology
• * Bioinformatics
• * Virology

Background:

• * Human endogenous retroviruses (HERVs) are implicated as cofactors in chronic diseases like cancer and autoimmune disorders.
• * The immunosuppressive domain (ISD) within retroviral envelope proteins is a key pathogenic element, but few HERV ISDs are annotated.
• * Existing prediction software lacks sufficient accuracy for identifying specific HERV ISDs.

Purpose of the Study:

• * To develop a computational model for accurate identification of ISDs in HERVs using only genomic sequences.
• * To expand the annotation of HERV ISDs and provide a tool for their prediction.

Main Methods:

• * Development of a computational model based on genome sequences for ISD prediction.
• * Validation using a Jack-knife test, achieving 97.9% classification accuracy.
• * Application of the model to scan 117 HERV families.

Main Results:

• * Prediction and annotation of 1002 novel putative ISDs across human chromosomes.
• * Demonstrated applicability of the model to other retroviruses like HTLV, STLV, and MLV.
• * Development of user-friendly software, ISDTool, for model application.

Conclusions:

• * The developed computational model offers high accuracy for identifying HERV ISDs.
• * This work significantly expands the known repertoire of HERV ISDs, aiding disease research.
• * ISDTool provides a valuable resource for researchers studying HERVs and related viruses.