This is the code repository of the TriGORANK model : A Gene Ontology Enriched Learning-to-Rank Framework for Trigenic Fitness Prediction.
Description: We developed machine learning models and learning to rank models for recommmending high-fitness triplet gene mutants for wet-lab experiments. We utilized gene ontology to obtain graph representations of triplet genes, gene similarities, gene intersection, path based features to add to the learning to rank model
Labhishetty et al., In Proceedings of IEE BIBM 2021,TriGORank: A Gene Ontology Enriched Learning-to-Rank Framework for Trigenic Fitness Prediction. https://doi.org/10.1109/bibm52615.2021.9669503
learning_to_rank_onto_features.py : learning to rank model with added ontology based features .
run_script.sh : For training models and generating results with different top-k relevant and precision settings for training and testing.
If you use our work in your research please cite:
@INPROCEEDINGS{9669503,
author={Labhishetty, Sahiti and Lourentzou, Ismini and Volk, Michael Jeffrey and Mishra, Shekhar and Zhao, Huimin and Zhai, Chengxiang},
booktitle={2021 IEEE International Conference on Bioinformatics and Biomedicine (BIBM)},
title={TriGORank: A Gene Ontology Enriched Learning-to-Rank Framework for Trigenic Fitness Prediction},
year={2021},
volume={},
number={},
pages={1841-1848},
doi={10.1109/BIBM52615.2021.9669503}}
By using this source code you agree to the license described in https://github.com/sahitilucky/TriGORank/blob/master/LICENSE