An expanded evaluation of protein function prediction methods shows an improvement in accuracy
Abstract
Background: The increasing volume and variety of genotypic and phenotypic data is a major defining characteristic of modern biomedical sciences. At the same time, the limitations in technology for generating data and the inherently stochastic nature of biomolecular events have led to the discrepancy between the volume of data and the amount of knowledge gleaned from it. A major bottleneck in our ability to understand the molecular underpinnings of life is the assignment of function to biological macromolecules, especially proteins. While molecular experiments provide the most reliable annotation of proteins, their relatively low throughput and restricted purview have led to an increasing role for computational function prediction. However, accurately assessing methods for protein function prediction and tracking progress in the field remain challenging. Methodology: We have conducted the second Critical Assessment of Functional Annotation (CAFA), a timed challenge to assess computational methods that automatically assign protein function. One hundred twenty-six methods from 56 research groups were evaluated for their ability to predict biological functions using the Gene Ontology and gene-disease associations using the Human Phenotype Ontology on a set of 3,681 proteins from 18 species. CAFA2 featured significantly expanded analysis compared with CAFA1, with regards to data set size, variety, and assessment metrics. To review progress in the field, the analysis also compared the best methods participating in CAFA1 to those of CAFA2. Conclusions: The top performing methods in CAFA2 outperformed the best methods from CAFA1, demonstrating that computational function prediction is improving. This increased accuracy can be attributed to the combined effect of the growing number of experimental annotations and improved methods for function prediction.
Cite
@article{arxiv.1601.00891,
title = {An expanded evaluation of protein function prediction methods shows an improvement in accuracy},
author = {Yuxiang Jiang and Tal Ronnen Oron and Wyatt T Clark and Asma R Bankapur and Daniel D'Andrea and Rosalba Lepore and Christopher S Funk and Indika Kahanda and Karin M Verspoor and Asa Ben-Hur and Emily Koo and Duncan Penfold-Brown and Dennis Shasha and Noah Youngs and Richard Bonneau and Alexandra Lin and Sayed ME Sahraeian and Pier Luigi Martelli and Giuseppe Profiti and Rita Casadio and Renzhi Cao and Zhaolong Zhong and Jianlin Cheng and Adrian Altenhoff and Nives Skunca and Christophe Dessimoz and Tunca Dogan and Kai Hakala and Suwisa Kaewphan and Farrokh Mehryary and Tapio Salakoski and Filip Ginter and Hai Fang and Ben Smithers and Matt Oates and Julian Gough and Petri Törönen and Patrik Koskinen and Liisa Holm and Ching-Tai Chen and Wen-Lian Hsu and Kevin Bryson and Domenico Cozzetto and Federico Minneci and David T Jones and Samuel Chapman and Dukka B K. C. and Ishita K Khan and Daisuke Kihara and Dan Ofer and Nadav Rappoport and Amos Stern and Elena Cibrian-Uhalte and Paul Denny and Rebecca E Foulger and Reija Hieta and Duncan Legge and Ruth C Lovering and Michele Magrane and Anna N Melidoni and Prudence Mutowo-Meullenet and Klemens Pichler and Aleksandra Shypitsyna and Biao Li and Pooya Zakeri and Sarah ElShal and Léon-Charles Tranchevent and Sayoni Das and Natalie L Dawson and David Lee and Jonathan G Lees and Ian Sillitoe and Prajwal Bhat and Tamás Nepusz and Alfonso E Romero and Rajkumar Sasidharan and Haixuan Yang and Alberto Paccanaro and Jesse Gillis and Adriana E Sedeño-Cortés and Paul Pavlidis and Shou Feng and Juan M Cejuela and Tatyana Goldberg and Tobias Hamp and Lothar Richter and Asaf Salamov and Toni Gabaldon and Marina Marcet-Houben and Fran Supek and Qingtian Gong and Wei Ning and Yuanpeng Zhou and Weidong Tian and Marco Falda and Paolo Fontana and Enrico Lavezzo and Stefano Toppo and Carlo Ferrari and Manuel Giollo and Damiano Piovesan and Silvio Tosatto and Angela del Pozo and José M Fernández and Paolo Maietta and Alfonso Valencia and Michael L Tress and Alfredo Benso and Stefano Di Carlo and Gianfranco Politano and Alessandro Savino and Hafeez Ur Rehman and Matteo Re and Marco Mesiti and Giorgio Valentini and Joachim W Bargsten and Aalt DJ van Dijk and Branislava Gemovic and Sanja Glisic and Vladmir Perovic and Veljko Veljkovic and Nevena Veljkovic and Danillo C Almeida-e-Silva and Ricardo ZN Vencio and Malvika Sharan and Jörg Vogel and Lakesh Kansakar and Shanshan Zhang and Slobodan Vucetic and Zheng Wang and Michael JE Sternberg and Mark N Wass and Rachael P Huntley and Maria J Martin and Claire O'Donovan and Peter N Robinson and Yves Moreau and Anna Tramontano and Patricia C Babbitt and Steven E Brenner and Michal Linial and Christine A Orengo and Burkhard Rost and Casey S Greene and Sean D Mooney and Iddo Friedberg and Predrag Radivojac},
journal= {arXiv preprint arXiv:1601.00891},
year = {2016}
}
Comments
Submitted to Genome Biology