Back

PSEA: A phenotypic similarity ensemble approach for prioritizes candidate genes to aid mendelian disease diagnosis

Wang, Z.; Liu, L.; Chen, C.; Liu, X.; Tang, F.; Zhang, Y.; Chen, Y.; Wang, Y.; Sun, J.; Peng, Z.

2021-10-15 bioinformatics
10.1101/2021.10.13.464308 bioRxiv
Show abstract

MotivationNext-generation sequencing (NGS) is increasingly applied to the molecular diagnosis of genetic disorders. However, challenges for the interpretation of NGS data remain given the massive number of variants produced by NGS. Careful assessment is required to identify the most likely disease-causing variants that best match the patients clinical phenotypes, which is highly experience-dependent and of low cost-effectiveness. ResultsThe human phenotype ontology (HPO) together with the information content (IC) are widely used for phenotypic similarity evaluation. Here, we introduce PSEA, a new phenotypic similarity evaluation tool capable of quantifying groups of HPO terms unbiasedly. By comparing with other methods, PSEA show optimal performance and show a higher tolerance to phenotypic noise or incompleteness. We also developed a web server for disease-causing gene prioritization and HPO-gene weighted linkage visualization. AvailabilitySource code and Web service are free available at https://github.com/zhonghua-wang/psea and https://phoenix.bgi.com/psea, respectively. Contactwangzhonghua@genomics.cn Supplementary informationSupplementary data are available at Bioinformatics online.

Matching journals

The top 5 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.