Reproduce “ARO: Aligned Representation learning for multi-Omics data” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
Amogh Singh, Yash Shah, Chiara D'Ercoli, Arash Mehrjou, Patrick Schwab, Timothy Jones, Pietro Liò
ARO addresses the gap between available molecular measurements and complete multi-omics profiles by using one shared autoencoder to reconstruct masked features, including non-transcriptomic features from transcriptomics. On normalized data, ARO MASKED reaches test MSE 0.14, matching PCA; embeddings from ARO UNMASKED yield 100% test cancer/non-cancer accuracy versus 72.41% for PCA, in a small cohort without external validation.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.