Exploring the unknown to better appreciate the 'known': a bioinformatics odyssey ...

Mass spec-based proteomics relies on knowing what to look for when identifying peptides (bottom-up) or proteoforms (top-down), notwithstanding de novo efforts. These search spaces are defined by sequence collections of proteins (fasta), and despite the constant churn of updates from different data producers, we accept that, at least for humans, we “know” what should be in a sample and everyone agrees what we should use. In contrast, working in non-model systems requires an appreciation of the unknown and a constant questioning of any species-specific resource that comes online. This healthy skepticism for search space may likewise be warranted in the human proteomics. To demonstrate these concepts, we will delve into the trials and tribulations faced when analyzing non-model organisms, from crows to sea lions, including misappropriated fasta from other species, and how search space choices affect results. These same lessons will be re-hashed using human pangenomes, with an eye to population-level proteomics. ...