isomiRTar

isomiRTar catalogs and analyzes 5'-end microRNA isoforms (5'-isomiRs) arising from imprecise Drosha and Dicer cleavage and predicts and validates their targetomes across TCGA pan-cancer sequencing cohorts to study isomiR-mediated gene regulation.

Key Features:

• Extensive Database: Compiles 1022 5'-isomiRs identified in 9282 TCGA tumor samples across 31 cancer types.
• Target Prediction: Uses miRDB and TargetScan to predict potential target genes for each 5'-isomiR sequence.
• Co-expression Analysis: Validates predicted isomiR–target pairs by assessing significant negative correlations in expression across cancer types.
• Expression Landscape Insights: Reports cancer-specific and conserved 5'-isomiR expression profiles across diverse cancers.
• Targetome Overlap Analysis: Compares targetomes of isomiR pairs differing by a single 5'-nucleotide and quantifies overlap using the Jaccard index (median 0.06).

Scientific Applications:

• Cancer Research: Enables investigation of isomiR-mediated gene silencing mechanisms and discovery of candidate biomarkers and regulatory interactions across cancer types.
• Tumor Suppressor Identification: Supports identification of candidate tumor-suppressor isomiRs, exemplified by hsa-miR-203a-3p-+1, hsa-miR-192-5p-+1, and hsa-miR-148a-3p-0 in colorectal cancer.

Methodology:

Uses TCGA high-throughput sequencing data to catalog 5'-isomiRs, predicts targets with miRDB and TargetScan, validates pairs via co-expression analysis of negative correlations, and quantifies targetome overlap with the Jaccard index.