A vast majority of the genome annotation of "protein coding genes" is based on experimental data collected on the transcript level. While powerful, the analysis of some proteins will only be possible at the protein level. Mass spectrometry is a powerful experimental tool that can contribute significantly to the annotation of genomes at the protein level. Combining mass spectrometry with molecular weight fractionation techniques strengthens our ability to identify more proteins at varying abundances and can resolve different protein isoforms. In addition, analyzing different developmental stages of an organism will improve our knowledge of changes of protein abundance during development. We are currently analyzing 12 developmental stages of C. elegans - N2 mixed embryo, N2 L1, N2 L2, N2 L3, N2 L4, N2 YA, N2 dauer, spe-9 L4, spe-9 YA, spe-9 adult, spe-9 adult reproductive diapause (ARD) and him- 8. The developmental stages are separated into 16 molecular weight fractions ranging from 3.5 - 500 kDa using the GelFree 8100 fractionation system (Protein Discoveries). An unfractionated set is also analyzed for all developmental stages. The peptides from each fraction are analyzed using a 120-minute LC-MS/MS run on a Thermo LTQ-Orbi-Velos mass spectrometer. A biological and analytical replicate is performed for each sample. The MS/MS spectra are assigned to peptides using the SEQUEST algorithm, q-values are assigned at the peptide level using Percolator and the peptides are assembled into proteins using an in-house implementation of IDPicker. Presently, we have analyzed the unfractionated samples and the half of the first molecular weight fraction (3.5-18 kDa). We identified 2736 non-redundant proteins and 14558 peptides in the unfractionated sample set and 1828 non-redundant proteins and 7122 peptides in the first molecular weight fraction at a very stringent peptide q-value cut-off of 0.01. Extrapolation of the current results out to all biochemical fractions suggests we will have sufficient depth of coverage to provide quantitative measures of protein abundance and confirmation of protein coding genes for approximately half the protein coding genes in the genome. Additionally, because we have data about the molecular weight fraction that the tryptic peptides were identified in, we will show examples of proteolytic processing of the protein coding gene to a smaller functional protein.

Affiliations:
- Dept Genome Sciences, Univ Washington, Seattle, WA.
- Fred Hutchinson Cancer Research Center, Seattle, WA