TaxTriage

TaxTriage identifies putative pathogens from short and long-read DNA and RNA metagenomic sequencing data through read classification, mapping, and de novo assembly combined with comparisons to curated pathogens and abundance expectations from healthy cohort data.

Key Features:

• Read Classification: Classifies sequencing reads to identify taxonomic origin.
• Mapping: Maps reads to reference sequences for pathogen identification.
• De Novo Assembly: Assembles reads without reference to detect novel or divergent pathogens.
• Curated Pathogen Comparison: Identifies putative pathogens through comparisons to curated pathogen databases.
• Healthy Cohort Abundance Expectations: Uses abundance data from healthy cohorts to distinguish true pathogens from background or commensal organisms.
• Custom Confidence Score: Generates a custom confidence score for reported organisms.
• Organism Discovery Report: Compiles final analysis summaries listing likely pathogens and supporting data.
• Host Species Customization: Incorporates customization options to improve performance for host species of interest in public health and veterinary diagnostics.
• Nextflow Workflow: Implements the pipeline using Nextflow for flexible deployment.

Scientific Applications:

• Clinical Pathogen Detection: Detects expected pathogens and co-infections from clinical metagenomic sequencing data with comparable sensitivity and increased specificity relative to alternative cloud-based processing pipelines.
• Outbreak Investigation: Evaluates outbreak datasets for pathogen identification.
• In Silico Evaluation: Supports performance benchmarking using published in silico datasets.
• Veterinary Diagnostics: Enables customized pathogen detection for non-human host species.
• Public Health Surveillance: Supports pathogen identification for public health applications.

Methodology:

TaxTriage combines read classification, mapping, and de novo assembly approaches. Putative pathogens are identified through comparisons to curated pathogens and abundance expectations from healthy cohort data. The pipeline is implemented in Nextflow.