Fig. 1. The K-means cluster analysis of pX01 gene expression data over nine culture conditions. Clustering was performed using K=7 and 100 maximum cycles. The toxin genes pagA, lef and cya cluster with the gerX operon genes, pagR and 16 other transcripts.

Fig. 2. Microarray mini-sequencing reaction time course. Microarray mini-sequencing reactions were allowed to proceed for 5, 15, 30, 60 and 120 min. Each reaction in the time course used template derived from the same PCR reaction. (A) microarray images were collected for all time points using identical scanner and PMT settings. Relevant sub-regions of the microarrays for 5, 30 and 60 min time points are shown without adjustment of contrast or brightness to allow visual comparison of signal intensity. (B) Average signal to noise values were calculated, as described it the text, for all sites queried in the mini-sequencing reaction. Values were plotted for all examined time points. The highest signal to noise values were obtained during the first hour of reaction. At later time points increased background fluorescence was apparent and a concomitant decrease in signal to noise can be seen for data from these longer incubations. Accurate genotyping data was obtained from all time points.

Table 1. Detection primer sequences

Detection primers were synthesized for each polymorphism to be distinguished (Base). All oligonucleotides carried a 5′-amine group for coupling to the microarray substrate followed by a (T)₉ spacer (not shown). The 3′- most base was positioned at the SNP site to be queried. Position designations are based on GenBank accession number M22589.

Table 2. Expression ratios and annotations for genes within the toxin gene cluster

Genes that fall within the cluster containing the toxin (cya, lef, pagA) and germination (gerXCAB) genes are shown. Annotations are from Okinaka et al. (1999) and Ariel et al. (2002). All gene expression ratios are relative to the 28 °C early growth condition. All shown genes exhibit expression ratios significantly different from 1.0 by a one-tailed t-test (P<0.05) at 37 °C + CO₂ mid- and/or late-phase growth as indicated by underlining in the appropriate column. Genes located within the PAI are shaded.

Table 3. Microarray-based mini-sequencing of pagA SNPs

Microarray-based mini-sequencing of seven B. anthracis strains found by traditional automated sequencing to exhibit genotype diversity at one or more queried pagA sites. Positions 1998, 2883, 3481, 3496, and 3602 were previously described as polymorphic by Price et al. (1999). Position 2784 was identified during the sequencing of the 90 B. anthracis strains characterized here. Microarray-based sequencing data agreed with traditional automated sequencing at all queried sites. All SNP position numbers are based on GenBank accession number M22589.