Robust evaluation of deep learning-based representation methods for survival and gene essentiality prediction on bulk RNA-seq data

• Published in: Scientific reports Vol. 14; no. 1; pp. 17064 - 15
• Main Authors: Gross, Baptiste, Dauvin, Antonin, Cabeli, Vincent, Kmetzsch, Virgilio, El Khoury, Jean, Dissez, Gaëtan, Ouardini, Khalil, Grouard, Simon, Davi, Alec, Loeb, Regis, Esposito, Christian, Hulot, Louis, Ghermi, Ridouane, Blum, Michael, Darhi, Yannis, Durand, Eric Y., Romagnoni, Alberto
• Format: Journal Article
• Language: English
• Published: London Nature Publishing Group UK 24-07-2024; Nature Publishing Group; Nature Portfolio
• Subjects: 631/114; 631/114/1305; 631/114/1314; 631/114/2163; 631/67; Benchmarking; Cell culture; Computational Biology - methods; Deep Learning; Gene essentiality; Genes, Essential; Humanities and Social Sciences; Humans; Medical research; multidisciplinary; Neoplasms - genetics; Neoplasms - mortality; Performance evaluation; Predictions; Representation learning; RNA-Seq - methods; RNAseq; Science; Science (multidisciplinary); Survival; Survival prediction; Training; Representation learning; Deep learning; Survival prediction; Benchmarking; RNAseq; Gene essentiality
• DOI: 10.1038/s41598-024-67023-8
• EISSN: 2045-2322

Deep learning (DL) has shown potential to provide powerful representations of bulk RNA-seq data in cancer research. However, there is no consensus regarding the impact of design choices of DL approaches on the performance of the learned representation, including the model architecture, the training methodology and the various hyperparameters. To address this problem, we evaluate the performance of various design choices of DL representation learning methods using TCGA and DepMap pan-cancer datasets and assess their predictive power for survival and gene essentiality predictions. We demonstrate that baseline methods achieve comparable or superior performance compared to more complex models on survival predictions tasks. DL representation methods, however, are the most efficient to predict the gene essentiality of cell lines. We show that auto-encoders (AE) are consistently improved by techniques such as masking and multi-head training. Our results suggest that the impact of DL representations and of pretraining are highly task- and architecture-dependent, highlighting the need for adopting rigorous evaluation guidelines. These guidelines for robust evaluation are implemented in a pipeline made available to the research community.