RNAlign: Alignment of tumor and cell line transcriptomes using conditional VAEs

Jacob Alvarez, Kiran Krishnamachari, Anders Jacobsen Skanderup

Keywords: generative models, gene expression, cancer models, cancer cell lines, preclinical cancer models, tumor samples, stromal cell contamination, in-vitro adaptation, biomarkers, conditional variational auto-encoders, CVAE, gene expression profiles, pan-cancer, alignment, batch effects, clinical precision medicine, machine learning, bioinformatics, genomic data, translational cancer research, regularization techniques, cancer research workflows, transcriptomic harmonization, tumor transcriptomes, model comparison

TL;DR: We perform best-in-class alignment of cell line and tumor data using a novel CVAE framework.

Confirmation Of Submission Requirements: I submit an abstract. It uses the template provided on the submission page and is no longer than 2 pages.

Submission Number: 125