We too could produce -- as any talented group of bioininformaticsts could produce -- a poster (never mind marketing material) that showcases "more peptides identified." Even if it were not misleading -- there are many, many ways in which one tool in a specific circumstance could produce more or less identifications than any other tool -- it would be besides the point.
What if, instead of showing "more peptides in Excel sheet," we could help you identify the specific panel of analytes that actually differentiate cohorts, even though those analytes plus modifications may not be present in the FASTA file plus searched conditions, eg unexpected glycosolated forms only present in a small percentage of the population but that ultimately explains why some group of individuals are susceptible to a disease while the vast majority of others are not?
What we believe we are achieving with the combination of a novel protocol (gs-DIA) and corresponding informatics supser-structure (Deepsearch) is a simple-to-explain protocol in which we can accurately quantify almost every analyte present in the MS, even without a FASTA file being present or the modifications being known apriori.
Why is this so critical?
This may be somewhat obvious to you, the reader, (both intuitively and based on your years of reading biology papers etc), but if, deep-down, the primary rainson d'etre of an over $10k discovery-MS project is to differentiate samples, then novel proteins/peptides that differentiate samples (the easy ones have already been taken) are almost guaranteed to NOT be in a simple list of unmodified peptides present in the public FASTA files. Instead, not only would they be 100x or even 1000x more likely be a modified peptide, but they are likely going to be a rare, unexpected or unknown (eg present only in a subset of the 'disease' population) glycoform. So no amount of "my tool has 25% more ids than your tool" will ever make any effective difference to your multi-million dollar MS and corresponding discovery projects.
But, if we can quantify almost every single analyte in the MS without the analyte necessarily being present in the FASTA file or the user knowing apriori what possible modification (of the 100s/1000s of potential ones) it could be, then we have a very real chance of changing science, publishing those 'seminal' papers, and, if you're on the commercial side (or even a member of a university with an active tech transfer office), being the first to make relevant patentable discoveries.
That's what we believe Deepsearch can do. We would love to discuss (through a super brief phone call, max 5 minutes), how we could collaborate on your area of specialty.
Comments
Post a Comment