METATRYP v 2.0: Metaproteomic Least Common Ancestor Analysis for Taxonomic Inference Using Specialized Sequence Assemblies—Standalone Software and Web Servers for Marine Microorganisms and Coronaviruses

Abstract

We present METATRYP version 2 software that identifies shared peptides across the predicted proteomes of organisms within environmental metaproteomics studies to enable accurate taxonomic attribution of peptides during protein inference. Improvements include ingestion of complex sequence assembly data categories (metagenomic and metatranscriptomic assemblies, single cell amplified genomes, and metagenome assembled genomes), prediction of the least common ancestor (LCA) for a peptide shared across multiple organisms, increased performance through updates to the backend architecture, and development of a web portal ( ). Major expansion of the marine METATRYP database with predicted proteomes from environmental sequencing confirms a low occurrence of shared tryptic peptides among disparate marine microorganisms, implying tractability for targeted metaproteomics. METATRYP was designed to facilitate ocean metaproteomics and has been integrated into the Ocean Protein Portal ( ); however, it can be readily applied to other domains. We describe the rapid deployment of a coronavirus-specific web portal ( ) to aid in use of proteomics on coronavirus research during the ongoing pandemic. A coronavirus-focused METATRYP database identified potential SARS-CoV-2 peptide biomarkers and indicated very few shared tryptic peptides between SARS-CoV-2 and other disparate taxa analyzed, sharing <1% peptides with taxa outside of the betacoronavirus group, establishing that taxonomic specificity is achievable using tryptic peptide-based proteomic diagnostic approaches.

Development of an Ocean Protein Portal for Interactive Discovery and Education

“…This is a key question within the field of metaproteomics because of the possibility that peptide constituents of proteins could be found in multiple organisms present within an individual sample. To address this, the OPP utilizes the software tool METATRYP we previously created that searches a database of all tryptic peptides among a group of organisms specified or within meta-omic assemblies from the environment. , METATRYP then identifies peptides that are shared among multiple organisms and reports which organisms share the peptides and calculates the “Least Common Ancestor” (LCA) of the identified taxa possessing this peptide. The METATRYP databases use the NCBI Taxonomy database to identify the ancestral phylogeny of the taxa identified that possess the peptide in question.…”

“…The results can be visualized in heatmap and tree formats to allow the user to gain an immediate understanding of who the Least Common Ancestor of the protein constituent is and their associated taxonomic lineages (Figure ). The OPP is currently using METATRYP Version 2 that has improved performance, can calculate the Least Common Ancestor, and has the capacity to separate its database into genomes, metagenomes, and metagenomic products that are described in a separate manuscript …”

“…The OPP Version 1.1 is currently built using an Elasticsearch database for protein and peptide data, that is accessed by the UI, generated with Django, Javascript, OceansMap, Bokeh, and Matplotlib code. , METATRYP Version 2 Least Common Ancestor software uses PostgreSQL and Python . Ingestion of data occurs through a process where data generators deposit data for the three file types described in Table according to specified data templates while working with a BCO-DMO data manager.…”

A rigorous evaluation of optimal peptide targets for MS-based clinical diagnostics of Coronavirus Disease 2019 (COVID-19)

“…The workflow also included additional, optional in-line characterization of these peptides by searching against NCBI-non redundant (nr) BLAST-P and Unipept [ 40 ] analysis. Further offline analysis was performed using NCBI BLAST-P analysis as well as the MetaTRYP [ 34 ] coronavirus database. The peptide validation workflow can be found at COVID Galaxy website ( https://COVID19.galaxyproject.org/proteomics ).…”

“…We then interrogated these peptides using the PepQuery search engine [ 31 ] to confirm the quality of these PSMs and determine whether the matched sequences were unique to SARS-CoV-2 or could be better ascribed to the human proteome or that of another closely related coronavirus. Peptides and their associated PSMS which survived this rigorous filtering were then manually validated using the Multi-omics Visualization Platform [ 32 ] and further analyzed for specificity to the SARS-CoV-2 virus via BLAST-P [ 33 ] and MetaTryp [ 34 ]. Taken together, our analyses enable the construction of a high-confidence target peptide list that would form the basis of a targeted clinical proteomics assay for SARS-CoV-2 infection.…”