Refining Transcriptome Gene Catalogs by MS-Validation of Expressed Proteins

Area of Science:

• Proteomics
• Molecular Biology
• Bioinformatics

Background:

• Tandem mass spectrometry (LC-MS/MS) identifies proteins in complex mixtures, aiding biological function studies.
• Transcriptomes are crucial for non-model organisms, serving as gene catalogs and revealing metabolic potential.
• Determining the correct protein-coding reading frame is essential for accurate biological interpretation.

Purpose of the Study:

• To identify the actively translated protein-coding reading frames within the transcriptome of the dinoflagellate Lingulodinium polyedra.
• To assess the impact of using MS-validated protein sequences versus nucleic acid sequences on downstream bioinformatics analyses.
• To compare MS-validated protein sequences with longest open reading frame (ORF) datasets and evaluate protein-mRNA abundance correlations.

Main Methods:

• Combined LC-MS/MS data from multiple experiments on Lingulodinium polyedra extracts.
• Utilized a 74,655-sequence transcriptome to identify peptide matches.
• Compiled a dataset of 6,628 MS-validated protein sequences and compared them with their parental nucleic acid sequences using BLASTp and BLASTx.

Main Results:

• MS-validated protein sequences showed differences in gene ontology, BLAST hits, and KEGG pathway enzyme content compared to DNA sequence BLASTx analyses.
• MS-validated protein sequences differed from longest open reading frame (ORF) datasets.
• A poor correlation was observed between protein and mRNA abundance levels, a novel finding for dinoflagellates.

Conclusions:

• MS-validated protein sequence datasets provide a more accurate representation of cellular capacity than nucleic acid sequence datasets.
• Using MS-validated protein sequences can enhance the accuracy of downstream analyses in proteomics and transcriptomics.
• Developing MS-validated protein datasets can accelerate the interpretation of MS-MS spectra in proteomics.