Abstract
In this article, we provide a comprehensive study of the content of the Universal Protein Resource (UniProt) protein data sets for human and mouse. The tryptic search spaces of the UniProtKB (UniProt knowledgebase) complete proteome sets were compared with other data sets from UniProtKB and with the corresponding International Protein Index, reference sequence, Ensembl, and UniRef100 (where UniRef is UniProt reference clusters) organism-specific data sets. All protein forms annotated in UniProtKB (both the canonical sequences and isoforms) were evaluated in this study. In addition, natural and disease-associated amino acid variants annotated in UniProtKB were included in the evaluation. The peptide unicity was also evaluated for each data set. Furthermore, the peptide information in the UniProtKB data sets was also compared against the available peptide-level identifications in the main MS-based proteomics repositories. Identifying the peptides observed in these repositories is an important resource of information for protein databases as they provide supporting evidence for the existence of otherwise predicted proteins. Likewise, the repositories could use the information available in UniProtKB to direct reprocessing efforts on specific sets of peptides/proteins of interest. In summary, we provide comprehensive information about the different organism-specific sequence data sets available from UniProt, together with the pros and cons for each, in terms of search space for MS-based bottom-up proteomics workflows. The aim of the analysis is to provide a clear view of the tryptic search space of UniProt and other protein databases to enable scientists to select those most appropriate for their purposes.
Highlights
Most of the current MS-based bottom-up proteomics workflows make use of collections of sequences to match peptide sequences to experimental spectra and to infer the proteins to which those peptides belong [1]
UPI data sets contain additional peptide-level sequence information compared to the CPI data sets for human and mouse (Table 4; see Supporting Information Notes “Universal Protein Resource (UniProt) complete proteomes and other sequences”). (iii) If variation data need to be considered in a given study, a variant-expanded data set is the proper choice
An additional choice consists in focusing on a subset of variation considering only the ones directly linked to disease, obtaining a data set focusing on detrimental variations with a lower sequence redundancy in the variant-expanded sequence data set
Summary
Most of the current MS-based bottom-up proteomics workflows make use of collections of sequences Uniprot.org) [2] is among the most used protein sequence and functional annotation providers. Among the UniProt databases (DBs) are the UniProt knowledgebase (UniProtKB) that acts as the central hub for the collection of functional information on proteins and the UniProt reference clusters (UniRef) [3] that merge closely related sequences based on sequence identity. UniProtKB consists of two sections: UniProtKB/Swiss-Prot, which is manually annotated and reviewed, and UniProtKB/TrEMBL, which is automatically annotated and is unreviewed.
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
