Figure 1.
Frequency of common oligonucleotides in the genome of P. fluorescens SBW25.
Data shows comparisons to both a random model, and to the closely related P. fluorescens Pf0-1 genome. The random model is based on 100 genomes generated with the same dinucleotide content, replication bias and length as the SBW25 genome. P. fluorescens Pf0-1 shares the same GC-content as SBW25 and has a highly similar dinucleotide content (Table S1); coding density differs by 1.7% and the genome length differs by 4%.
Table 1.
Short repetitive sequence groups in the SBW25 genome.
Table 2.
Frequency of GI, GII, and GIII 16-mers in the extragenic space of the SBW25 genome.
Figure 2.
Frequency of next-neighbor distances for GI, GII, and GIII sequences in the genome of P. fluorescens SBW25.
Data are next-neighbor distances for 1,053 GI, GII and GIII sequences in extragenic space, compared to a random model (inset). The peaks at 71 and 110 bp correspond to doublets of GI and GII sequences, respectively. The peak at 184 bp corresponds to GI–GIII tandem repeat clusters (see text). No significant deviation from the random model was noted for next-neighbor distances above 200 bp. The next-neighbor distances of 16-mers randomly assigned to extragenic space is the average of 10,000 simulations (inset).
Table 3.
Frequency of REP clusters within the SBW25 genome.
Table 4.
Frequency of REP doublets within the SBW25 genome.
Figure 3.
Average pairwise identity of REP sequences found in singlets, doublets, and clusters.
Data are average pairwise identity of REPs found as singlets, doublets and clusters (clusters contain more than three REPs). Error bars show standard deviation. Statistical testing (jackknife) shows the average pairwise identity of 16-mers from REP doublets (and clusters for GI and GIII, P-value<1e-10) to be significantly greater than the average pairwise identity of 16-mers obtained from REP singlets: this is true for comparisons within each of the REP groups (P<1e-10 for GI; P<1e-8 for GII and GIII).
Figure 4.
General organization and predicted secondary structure of REPINs.
(A) Alignment of 101 GI REP doublets forming hairpins (REPINs) from SBW25 (37 are shown) shows a symmetrical (palindromic) organization comprised of two highly conserved regions separated by a spacer. Top line shows the consensus sequence followed by a graph displaying identity to the consensus (green denotes 100% identity). Two invariant regions of 16 bp are found in the left and right ends (LE, RE). These sequences are organized as inverted repeats and define the most abundant 16-mer in the SBW25 genome (black box). Each 16-mer overlaps a GI REP sequence (red box). (B) General REPIN features including LE and RE, each comprised of a highly conserved 16-mer (black) overlapping a REP sequence (red), with the two ends separated by a spacer. For a GI doublet the distance between the first residues of the two invariant 16-mers is 71 bp. Complementary bases permit formation of a hairpin structure (arrows). (C) Three excision events detected from Illumina sequencing reads reveal a putative transposition intermediate. Full-length sequences show three genomic regions located between 2,577,312–2,577,231, 3,857,520–3,857,439 and 5,683,545–5,683,624 bp on the SBW25 genome, each of which contains a REPIN. The partial sequences below each genomic region are Illumina reads from which the REPIN has been excised (see also Figure S6). (D) Cartoon of the excised region indicating putative transposition intermediate. Note the 5′-tail, which generates an asymmetrical sequence. (E) Secondary structure prediction for the consensus GI REPIN shows that the conserved bases on each side can pair resulting in a long hairpin (E, left). Predictions for transposition intermediates in the same order as the alignments in (C): the second, third and fourth hairpin correspond to the first, second and third alignment. The single stranded 5′-tail is free to pair with a complementary sequence.
Figure 5.
Proximity of GI, GII, and GIII REPIN clusters to RAYT genes in the P. fluorescens SBW25 genome.
The RAYT genes in SBW25 are pflu3939, yafM and pflu2165.
Figure 6.
REP singlet to doublet ratios for REP sequences from bacterial genomes.
Data are the most abundant 16-mers found within the flanking non-coding DNA of RAYT genes from 18 genomes. In order to include related 16-mers, a set of degenerate sequences was produced by allowing up to two substitutions per 16-mer.