Rxivist logo

Improving Deconvolution Methods in Biology through Open Innovation Competitions: An Application to the Connectivity Map

By Andrea Blasco, Ted Natoli, Michael G Endres, Rinat A Sergeev, Steven Randazzo, Jin H Paik, N. J. Maximilian Macaluso, Rajiv Narayan, Xiaodong Lu, David Peck, Karim R Lakhani, Aravind Subramanian

Posted 10 Jan 2020
bioRxiv DOI: 10.1101/2020.01.10.897363

Do machine learning methods improve standard deconvolution techniques for gene expression data? This paper uses a unique new dataset combined with an open innovation competition to evaluate a wide range of gene-expression deconvolution approaches developed by 294 competitors from 20 countries. The objective of the competition was to separate the expression of individual genes from composite measures of gene pairs. Outcomes were evaluated using direct measurements of single genes from the same samples. Results indicate that the winning algorithm based on random forest regression outperformed the other methods in terms of accuracy and reproducibility. More traditional gaussian-mixture methods performed well and tended to be faster. The best deep learning approach yielded outcomes slightly inferior to the above methods. We anticipate researchers in the field will find the dataset and algorithms developed in this study to be a powerful research tool for benchmarking their deconvolution methods and a useful resource for multiple applications. ### Competing Interest Statement The authors have declared no competing interest.

Download data

  • Downloaded 482 times
  • Download rankings, all-time:
    • Site-wide: 68,958
    • In genomics: 4,807
  • Year to date:
    • Site-wide: 58,492
  • Since beginning of last month:
    • Site-wide: 61,268

Altmetric data

Downloads over time

Distribution of downloads per paper, site-wide