FastViFi

FastViFi detects viral infection and integration sites of DNA viruses within host genomes and identifies human–viral hybrid transcripts from RNA-seq to characterize viral contributions to disease, particularly cancer.

Key Features:

• Viral integration detection: Detects viral infection sites and integration points within the human genome for DNA viruses.
• Two-stage filtering methodology: Implements a two-stage filtering methodology that reduces computational demands by approximately two orders of magnitude compared to traditional approaches.
• Ensemble hidden Markov models: Uses ensemble hidden Markov models (HMMs) tailored for specific viruses, including Epstein–Barr virus (EBV), Hepatitis B virus (HBV), Hepatitis C virus (HCV), and Human Papillomavirus (HPV).
• RNA-seq hybrid transcript detection: Analyzes RNA-seq data to detect human viral hybrid transcripts that fuse viral and human genes.
• Performance: Processes samples in under 10 minutes per sample in reported evaluations versus traditional methods that may exceed 40 CPU-hours per sample.
• Validation datasets: Demonstrated on simulated datasets and cancer genomic data.

Scientific Applications:

• Oncology: Elucidates the role of viral infections and integration in cancer etiology by identifying integration sites and hybrid transcripts.
• Transcriptomics of viral-host fusions: Identifies human–viral hybrid transcripts from RNA-seq to study mechanisms of oncoviral pathogenicity.
• Detection in variable viral genomes: Enables detection of diverse and highly variable viral genomes in genomic and transcriptomic datasets.

Methodology:

Computational methods explicitly include a novel two-stage filtering methodology and ensemble hidden Markov models (HMMs) tailored for EBV, HBV, HCV, and HPV, plus analysis of RNA-seq data to detect human–viral hybrid transcripts; evaluations used simulated and cancer genomic data and compared runtimes against traditional ensemble HMM approaches.