IMMUNOLOGY2026™ Conference Recordings For Attendees-TIDEPOOL, a method for inferring disease trajectories across cohorts, identifies neutrophil-driven trajectories associated with severity of infections

TIDEPOOL, a method for inferring disease trajectories across cohorts, identifies neutrophil-driven trajectories associated with severity of infections

Back to course

Video Transcription

Video Summary

The speaker presents Tidepool, a method that adapts single-cell trajectory inference to bulk transcriptomics to model disease progression as a continuum rather than forcing samples into simple healthy/sick or moderate/severe categories. Because many diseases are heterogeneous, Tidepool integrates many public RNA-seq and microarray datasets, conormalizing them using healthy controls and projecting samples into a low-dimensional trajectory space. Applied to over 2,000 infection samples from 24 datasets, it revealed two diverging trajectories from healthy to moderate and severe infection. Gene modules distinguishing these paths were linked to severity and largely came from neutrophils, especially immature and pro-inflammatory subsets in severe disease, while a protective interferon-associated module marked the less severe path. In independent sepsis/ARDS cohorts, these patterns replicated and improved severity prediction when moderate patients were split by trajectory. The work suggests trajectory-based analysis can identify high-risk moderate patients for earlier intervention.

Meta Tag

Date April 18, 2026 1:00 PM - 1:15 PM

Room 102

Session Computational Tool Development

Speaker Holly McCann

Track Computational and Systems Immunology (COMP)

Year 2026

Keywords

Tidepool

trajectory inference

bulk transcriptomics

disease progression

sepsis severity

April 18, 2026 1:00 PM - 1:15 PM

102

Computational Tool Development

Holly McCann

Computational and Systems Immunology (COMP)

2026