Back

SurfacOmics: an R shiny application integrating variable Feature Selection for gene biomarker discovery using Elastic-Net Regularization

Tripathi, S.; Kajla, P.; Abbasi, B. A.; Kota, K. P.; Bailey, A.; Varma, B.

2025-06-17 bioinformatics
10.1101/2025.06.12.659259 bioRxiv
Show abstract

Affordable sequencing technologies have resulted in a rapid rise in genomic and proteomic data. As a result, a massive amount of data is being analyzed, and the outcomes must be summarized as relevant clinical biomarkers. One of the major challenges is enabling wet-lab researchers to make meaningful inferences from the data, even in the absence of expertise in pipeline development and statistical training. We present a user-friendly R shiny application, SurfacOmics, which allows the user to perform biomarker identification via penalized regression algorithms. It also enables researchers to choose a crucial binary variable, such as treatment or group, from the study design metadata to anticipate potential biomarkers. We have introduced a novel concept of scoring scheme to characterize potential biomarkers for prediction, prioritization, and ranking. The tool integrates Gene Ontology information for sub-cellular localization together with a manually curated knowledgebase to provide valuable insights on biological processes, molecular functions and antibody resources, offering a comprehensive view in a single interface.

Matching journals

The top 5 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.