Regeneration-Likeness Scoring of Human Wound-Healing Genes Using Machine Learning

Overview

This project created machine learning models trained from axolotl regeneration data to identify which human genes that show patterns similar to those involved in axolotl regeneration. Two machine learning models were created, with the scores combined to create an overall ‘regeneration-likeness’ score for every gene in the input dataset. The input dataset was a human scarring dataset, in order to provide greater contrast between fibrotic (scarring) and more ‘regeneration-like’ genes, highlighting which genes show the strongest conserved signals/patterns and play the most vital roles in this context. Several of the top 10 highest scoring genes (most regeneration-like) were uncharacterized, meaning that there is very limited information currently known—suggesting that they may play more vital roles in healing biology or may be potential candidates for study in regenerative medicine.

Awards (2)

• Bronze Medal
• Selected for CWSF 2026

Competition history

• CWSF 2026 Digital Technology

Related projects

Closest projects by meaning, across every fair and year in the corpus.