Vibrio chromosome-specific families

Abstract

We have compared chromosome-specific genes in a set of 18 finished Vibrio genomes, and, in addition, also calculated the pan- and core-genomes from a data set of more than 250 draft Vibrio genome sequences. These genomes come from 9 known species and 2 unknown species. Within the finished chromosomes, we find a core set of 1269 encoded protein families for chromosome 1, and a core of 252 encoded protein families for chromosome 2. Many of these core proteins are also found in the draft genomes (although which chromosome they are located on is unknown.) Of the chromosome specific core protein families, 1169 and 153 are uniquely found in chromosomes 1 and 2, respectively. Gene ontology (GO) terms for each of the protein families were determined, and the different sets for each chromosome were compared. A total of 363 different "Molecular Function" GO categories were found for chromosome 1 specific protein families, and these include several broad activities: pyridoxine 5' phosphate synthetase, glucosylceramidase, heme transport, DNA ligase, amino acid binding, and ribosomal components; in contrast, chromosome 2 specific protein families have only 66 Molecular Function GO terms and include many membrane-associated activities, such as ion channels, transmembrane transporters, and electron transport chain proteins. Thus, it appears that whilst there are many "housekeeping systems" encoded in chromosome 1, there are far fewer core functions found in chromosome 2. However, the presence of many membrane-associated encoded proteins in chromosome 2 is surprising.

Keywords: Vibrio comparative genomics; Vibrio core-genome; Vibrio pan-genome; chromosome-specific genes; comparative genomics.

Figure 1

Predicted genome characteristics (A). Distribution of the number of contiguous pieces (B). Distribution of the protein number per genome (C). Distribution of the average protein coding gene length per genome (D). Number of predicted 16S rRNA sequences.

Figure 2

Vibrio chromosome comparison. Comparison was performed for set of 18 genomes. The blue and green square boxes represent chromosomes 1 and 2, respectively. The red-colored box in the middle of the figure indicates inter-chromosomal comparison of V. cholerae species, and the black-colored triangles highlight similarities within the same chromosome of the species.

Figure 3

GO term analysis in proteins shared by chromosomes 1 and 2. The distribution is shared both as percentage on the axis and the absolute number above the bar. The absolute number reflects the amount of GO IDs that were connected to the pathway. The color code is as follows: red is the biological process, green is the cellular component, and blue is the molecular function.

Figure 4

GO term analysis in protein coding genes shared within chromosome 1 and missing in the core of chromosome 2. The distribution is shared both as percentage on the axis and the absolute number above the bar. The absolute number shows the amount of GO IDs that were connected to the pathway. The color code is as follows: red is the biological process, green is the cellular component, and blue is the molecular function.

Figure 5

GO term analysis in protein coding genes shared within chromosome 2 and missing in the core of chromosome 1. The distribution is shared both as percentage on the axis and the absolute number above the bar. The absolute number shows the amount of GO IDs that were connected to the pathway. The color code is as follows: red is the biological process, green is the cellular component, and blue is the molecular function.

Figure 6

Pairwise interspecies-specific genome comparison for chromosome 1 (A) and chromosome 2 (B). Analysis included a single representation of 7 known and 2 unknown species. The resulting percentage shows the ratio between the amount of species-specific families and the size of the total proteome. On average, each species contained between 18 and 33% specific protein families. Color intensity indicates the level of specificity.

Figure 7

Annotation and length distribution of proteins within specific-proteomes in draft genomes of V. cholerae (A). Distribution of profiles by assignment source: PfamA, Superfamily, TIGRFAM, and CD-HIT clustering (B). Protein coding gene length distribution by each profile type.

Figure 8

GO term analysis in proteins, specific to V. cholerae draft genomes. Distribution is shared both as the percentage on the axis and the absolute number above the bar. The absolute number shows the amount of GO IDs that were connected to the pathway. The color code is as follows: red is the biological process, green is the cellular component, and blue is the molecular function.