Computational protocol: Comparative proteome analysis of propionate degradation by Syntrophobacter fumaroxidans in pure culture and in coculture with methanogens

Similar protocols

Protocol publication

[…] The obtained MS/MS spectra were processed with MaxQuant v. 1.5.2.8. Database with the protein sequences of S. fumaroxidans was downloaded from UniProt (http://www.uniprot.org). The protein database of Desulfovibrio desulfuricans strain G11 was downloaded from GenBank accession number CP023415. An additional dataset with protein sequences of common contaminants (trypsin, human keratins and bovine serum albumin) was included. False discovery rates (FDR) of < 1% were set at peptide and protein levels. Modifications for acetylation (Protein N‐term), deamidation (N, Q) and oxidation (M) were allowed to be used for protein identification and quantification. All other quantification settings were kept default. Filtering and further bioinformatics and statistical analysis were performed with Perseus v.1.5.3.0. Proteins included in our analysis contain at least two identified peptides of which at least one is unique and at least one unmodified. Reversed hits and contaminants were filtered out. Protein groups were filtered to require three valid values in at least one experimental group. Label‐free quantification (LFQ) intensities (values normalized with respect to the total amount of protein and all of its identified peptides) were used to analyse the abundance of proteins in the fractions and further statistical comparisons among conditions. LFQ intensities were transformed to logarithmic values base 10. Missing values were imputed with random numbers from a normal distribution, the mean and standard deviation of which were chosen to best simulate low abundance values close to noise level (Width: 0.3 and downshift 1.8 times). A multiple‐sample test (ANOVA) with permutation‐based FDR statistics (250 permutations, FDR = 0.01 and S0 = 1) was applied to filter significant proteins. PCA were performed with default settings and without category enrichment in components. Z‐score normalization in which the mean of each row (where each row is a protein in triplicate and in different conditions) is subtracted from each value and the result divided by the standard deviation of the row was applied before clustering. Hierarchical clustering of rows, using Euclidean distances, produced a heat map representation of the clustered data matrix. Row clusters were automatically defined (100) and exported to a new matrix. Imputed values were then replaced back to missing values and previously defined clusters were displayed in a new heat map. For D. desulfuricans the Z‐score and hierarchical clustering was done for columns instead of rows in order to compare the most abundant proteins detected in each condition. […]

Pipeline specifications

Software tools MaxQuant, Perseus
Application MS-based untargeted proteomics
Organisms Syntrophobacter fumaroxidans, Escherichia coli, Methanospirillum hungatei, Methanobacterium formicicum, Desulfovibrio desulfuricans
Chemicals Hydrogen, Succinic Acid