ISPRF

ISPRF predicts immune subtypes in clear cell renal cell carcinoma (ccRCC) to stratify tumors by immune-related gene expression and inform responses to immunotherapies such as anti-PD-1.

Key Features:

• Data integration: Integrates 831 ccRCC transcriptomic profiles aggregated from six distinct datasets.
• Unsupervised clustering: Clusters samples into immune subtypes based on immune cell enrichment scores.
• Hub gene identification (WGCNA): Uses weighted correlation network analysis to identify hub genes CTLA4, FOXP3, IFNG, and CD19 that distinguish subtypes and correlate with prognosis.
• Random forest classifier: Trains a random forest model on mRNA expression of the identified hub genes to predict immune subtype (ROC AUC = 0.78).

Scientific Applications:

• Personalized medicine: Identifies ccRCC patients with immune-enriched subtypes (e.g., subtype2) that are associated with improved immunotherapy responses.
• Prognostic assessment: Stratifies patients by immune subtype linked to differential prognosis.
• Research and development: Enables exploration of molecular immune mechanisms in ccRCC and supports discovery of biomarkers and therapeutic targets.

Methodology:

Integrates 831 ccRCC transcriptomes from six datasets, computes immune cell enrichment scores and applies unsupervised clustering to define immune subtypes, performs WGCNA to identify hub genes (CTLA4, FOXP3, IFNG, CD19), and trains a random forest classifier on hub-gene mRNA expression reporting ROC AUC = 0.78.