FloReMi

FloReMi performs survival analysis on flow cytometry datasets by identifying informative cell subsets and applying random survival forests to predict time-to-event outcomes such as progression to AIDS in HIV patients.

Key Features:

• Automated Data Preprocessing: Automates preprocessing of raw flow cytometry data to ensure data quality and consistency for downstream analysis.
• Identification and Selection of Informative Cell Subsets: Identifies and selects cell subsets informative for survival outcomes, emphasizing markers beyond CD4(+) T cell counts to capture early immunopathogenesis signals.
• Survival Regression Analysis: Implements survival regression using random survival forests to model censored time-to-event data and estimate progression rates to AIDS in HIV cohorts.

Scientific Applications:

• HIV progression modeling (FlowCAP IV): Predicts progression rate to AIDS in HIV patients from flow cytometry-derived cellular features, as demonstrated in the FlowCAP IV challenge.
• Biomedical survival analysis and longitudinal monitoring: Applies to survival analysis and longitudinal patient monitoring where flow cytometry-derived cellular biomarkers inform time-to-event models.

Methodology:

Automated preprocessing of raw flow cytometry data; clustering, classification, and visualization of cell populations; identification and selection of informative cell subsets; and survival regression using random survival forests for censored time-to-event data.