Published November 5, 2026 | Version v1.0.0

Specificity-controlled machine learning for neurodevelopmental disorder gene prioritization recovers recessive genes missed by constraint-based methods

  • 1. Department of Computer Science and Engineering, Keshav Memorial Institute of Technology, Hyderabad, Telangana 500029, India
  • 2. Department of Computer Science and Engineering, Keshav Memorial College of Engineering, Ibrahimpatnam, Telangana 501510, India
  • 3. Department of Computer Science and Engineering, Keshav Memorial Engineering College, Hyderabad, Telangana 500088, India

Description

A positive-unlabelled pipeline that ranks all 19,298 human protein-coding genes for neurodevelopmental-disorder risk, with specificity measured against genes causing Mendelian disease in unrelated organ systems, and prospective validation under strict temporal freezing of labels, protein network and publication counts. Benchmarked without retraining against forecASD, Krishnan et al., DAMAGES, DeepND and gnomAD LOEUF.

Notes

If you use this software or its outputs, please cite it as below.

Files

Advait-Sreekumar/ndd-gene-prioritization-v1.0.0.zip

Files (101.0 kB)

Name Size
md5:02c37f982cc036bb3aaea0aee64e18c7
101.0 kB Preview Download

Additional details