LITT

LITT identifies and ranks potential source cases within tuberculosis (TB) clusters by integrating whole-genome sequencing, clinical, and epidemiological data to support transmission chain analysis.

Key Features:

• Data integration: Integrates whole-genome sequencing, clinical, and epidemiological data to combine molecular and case-level information for cluster analysis.
• Source-case identification and ranking: Systematically identifies and ranks potential source cases within TB clusters.
• Automation of investigation logic: Automates procedures that mirror manual cluster-investigation methods to increase consistency and reproducibility.
• Empirical evaluation: Evaluated on 534 cases across 56 clusters (cluster sizes 2–69) from three U.S. jurisdictions, identifying the most likely source for 145 of 181 cases (80% concordance with human investigators).
• Data discrepancy detection: Discrepancies between LITT and investigators were often attributed to errors within the underlying dataset.

Scientific Applications:

• TB transmission chain analysis: Assists in reconstructing transmission links within TB clusters by combining genomic and epidemiological evidence.
• Cluster investigations: Provides a systematic framework for analyzing complex datasets during TB cluster investigations.
• Public health prioritization: Informs prioritization of resources and interventions to prevent further TB transmission.
• Decision support for field investigations: Supports interpretation of integrated genomic, clinical, and epidemiological data while not replacing detailed field investigations.

Methodology:

Integrates whole-genome sequencing, clinical, and epidemiological data and applies an algorithm that mirrors manual cluster-investigation processes to systematically identify and rank potential source cases.