Genomic Epidemiology of Clostridium botulinum Isolates from Temporally Related Cases of Infant Botulism in New South Wales, Australia.

Infant botulism is a potentially life-threatening paralytic disease that can be associated with prolonged morbidity if not rapidly diagnosed and treated. Four infants were diagnosed and treated for infant botulism in NSW, Australia, between May 2011 and August 2013. Despite the temporal relationship between the cases, there was no close geographical clustering or other epidemiological links. Clostridium botulinum isolates, three of which produced botulism neurotoxin serotype A (BoNT/A) and one BoNT serotype B (BoNT/B), were characterized using whole-genome sequencing (WGS). In silico multilocus sequence typing (MLST) found that two of the BoNT/A-producing isolates shared an identical novel sequence type, ST84. The other two isolates were single-locus variants of this sequence type (ST85 and ST86). All BoNT/A-producing isolates contained the same chromosomally integrated BoNT/A2 neurotoxin gene cluster. The BoNT/B-producing isolate carried a single plasmid-borne bont/B gene cluster, encoding BoNT subtype B6. Single nucleotide polymorphism (SNP)-based typing results corresponded well with MLST; however, the extra resolution provided by the whole-genome SNP comparisons showed that the isolates differed from each other by >3,500 SNPs. WGS analyses indicated that the four infant botulism cases were caused by genomically distinct strains of C. botulinum that were unlikely to have originated from a common environmental source. The isolates did, however, cluster together, compared with international isolates, suggesting that C. botulinum from environmental reservoirs throughout NSW have descended from a common ancestor. Analyses showed that the high resolution of WGS provided important phylogenetic information that would not be captured by standard seven-loci MLST.

A robust measure of correlation between two genes on a microarray.

BACKGROUND: The underlying goal of microarray experiments is to identify gene expression patterns across different experimental conditions. Genes that are contained in a particular pathway or that respond similarly to experimental conditions could be co-expressed and show similar patterns of expression on a microarray. Using any of a variety of clustering methods or gene network analyses we can partition genes of interest into groups, clusters, or modules based on measures of similarity. Typically, Pearson correlation is used to measure distance (or similarity) before implementing a clustering algorithm. Pearson correlation is quite susceptible to outliers, however, an unfortunate characteristic when dealing with microarray data (well known to be typically quite noisy.) RESULTS: We propose a resistant similarity metric based on Tukey's biweight estimate of multivariate scale and location. The resistant metric is simply the correlation obtained from a resistant covariance matrix of scale. We give results which demonstrate that our correlation metric is much more resistant than the Pearson correlation while being more efficient than other nonparametric measures of correlation (e.g., Spearman correlation.) Additionally, our method gives a systematic gene flagging procedure which is useful when dealing with large amounts of noisy data. CONCLUSION: When dealing with microarray data, which are known to be quite noisy, robust methods should be used. Specifically, robust distances, including the biweight correlation, should be used in clustering and gene network analysis.