surfaltr: An R/Bioconductor package to benchmark surface protein isoforms by rapid prediction and visualization of transmembrane topologies

Cell surface proteins form a major fraction of the druggable proteome and can be used for tissue-specific delivery of oligonucleotide/cell-based therapeutics. Surface protein isoforms are regulated by alternative splicing, which drives subcellular localization and transmembrane (TM) topology thereby shaping cell type specific signatures. Current advances in multiomic approaches have developed interest in discovery of tissue-specific alternatively spliced or novel surface protein isoforms. However, there exists a need for bioinformatic approaches for rapidly benchmarking the large number of isoforms identified by these approaches. To address this gap, we have developed, surfaltr, an R package which takes user input isoforms, pairs them with the known principal isoform of the gene, predicts TM topologies, and generates a customizable graphical output. Further, surfaltr facilitates prioritization of topologically diverse isoform pairs through incorporation of three different ranking metrics and through protein alignment functions. Here, we demonstrate the utility of our R package by evaluating the mouse retina-specific novel surface protein isoforms identified in Ray et al. 2020. surfaltr is freely available through Bioconductor (https://bioconductor.org/packages/surfaltr) and the vignette provides extensive instructions for implementation.