An evaluation for cross‐species proteomics research by publicly available expressed sequence tag database search using tandem mass spectral data

With 1383 tandem mass spectra derived from 120 individual protein spots separated by the two‐dimensional (2‐D) gel electrophoresis of protein samples from three different species, comparative analyses were performed by searching the Expressed Sequence Tag (EST) database (DB) and the NCBI non‐redunda...

Full description

Saved in:
Bibliographic Details
Published in:Rapid communications in mass spectrometry 2006-01, Vol.20 (18), p.2635-2640
Main Authors: Huang, Mei, Chen, Tong, Chan, ZhuLong
Format: Article
Language:English
Subjects:
Databases, Protein
Electrophoresis, Gel, Two-Dimensional - methods
Expressed Sequence Tags
Picea - chemistry
Picea - genetics
Pinus - chemistry
Pinus - genetics
Plant Proteins - analysis
Plant Proteins - chemistry
Plants - chemistry
Plants - genetics
Proteomics
Prunus - chemistry
Prunus - genetics
Species Specificity
Tandem Mass Spectrometry - methods
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:With 1383 tandem mass spectra derived from 120 individual protein spots separated by the two‐dimensional (2‐D) gel electrophoresis of protein samples from three different species, comparative analyses were performed by searching the Expressed Sequence Tag (EST) database (DB) and the NCBI non‐redundant (nr) DB of green plants, respectively, which uses the Mascot search engine to establish a statistical basis. It was confirmed that the former could identify more peptides manually validated by de novo sequencing (DNS) from fewer species in more closely phylogenetic relationships than the latter in a statistically significant manner. Our data demonstrated that correct peptide identifications were given low Mascot scores (e.g. 6–14) and incorrect peptide identifications were given high Mascot scores (e.g. 68–83). Our data also showed that the current evaluation approaches to protein assignments are unsatisfactory because a few ‘false‐positive’ proteins are recognized and several ‘false‐negative’ proteins are rescued by manual validation. Copyright © 2006 John Wiley & Sons, Ltd.
ISSN:0951-4198
1097-0231
DOI:10.1002/rcm.2631