Synapse
⌘+K
Synapse
PulseExploreClubsResearchersJournals
Instagram
HomeClubsExplore
December 8, 2025BloodOpen Access

Machine learning uncovers invariant evolutionary molecular trajectories in MDS.

View Full Paper
Ask AI
Bookmark
Share

Authors

ADArda DurmazCBCarlos Bravo‐PérezSPSimona Pagliuca

Discussion

Loading...

Member takes

Overview

Machine learning identifies invariant molecular clusters in MDS, suggesting clinical implications for patient progression dynamics.

Key Points

  • To explore how machine learning can define molecular clusters in MDS and their implications for disease progression.
  • Used autoencoder-based machine learning to analyze genetic features in patients with MDS.
  • Classified patients into molecular clusters (MCs) based on genetic similarities and investigated their transitions over time.
  • Analyzed a cohort of 3810 patients, with 320 studied serially to assess MC behavior during disease progression.
  • Identified distinct molecular clusters in MDS and their associations with disease progression dynamics.
  • 40% of patients with documented progression experienced MC reassignment, indicating non-random transitions.
  • MC13 emerged as the central node of malignant progression, affecting the clinical behavior of patients.

Cite This Study

Durmaz et al. (2025) studied this question.

synapsesocial.com/papers/69362f4b4fa91c937236d7b1https://doi.org/10.1182/blood-2025-2060
View Full Paper
Ask AI
Bookmark
Share