Camformer

Camformer is a convolutional neural network developed for the DREAM 2022 challenge "Predicting gene expression using millions of random promoter sequences". Our submission took 4th place and links to both the orignal model and our further work are provided below.

Follow-up work
GitHub: https://github.com/Bornelov-lab/Camformer
Reference: Dash T, Bornelöv S (2025) Predicting gene expression using millions of yeast promoters reveals cis-regulatory logic. bioRxiv 2025.01.25.634853.
Paper link: https://doi.org/10.1101/2025.01.25.634853

Official challenge paper
GitHub: https://github.com/de-Boer-Lab/random-promoter-dream-challenge-2022
Reference: Rafi AM, Nogina D, Penzar D, Lee D, Lee D, Kim N, Kim S, Shin Y, Kwek I-Y, Meshcheryakov G, Lando A, Zinkevich A, Kim B-C, Lee J, Kang T, Vaishnav ED, Yadollahpour P, Random Promoter DREAM Challenge Consortium (72 members, including Bornelöv S, Svensson F and Trapotsi M-A), Kim S, Albrecht J, Regev A, Gong W, Kulakovskiy IV, Meyer P, de Boer CG (2025) A community effort to optimize sequence-based deep learning models of gene regulation. Nature Biotechnology s41587-024-02414-w.
Paper link: https://doi.org/10.1038/s41587-024-02414-w
Challenge wiki: https://www.synapse.org/#!Synapse:syn28469146/wiki/617075

Original submission
GitHub: https://github.com/FredrikSvenssonUK/DREAM2022_Camformers
Team members: Bornelöv S, Svensson F, Trapotsi M-A