Whole-genome sequence (WGS) analysis has revolutionized the food safety industry by enabling high-resolution typing of foodborne bacteria. Higher resolving power allows investigators to identify origins of contamination during illness outbreaks and regulatory activities quickly and accurately. Government agencies and industry stakeholders worldwide are now analyzing WGS data routinely. Although researchers have published many studies that assess the efficacy of WGS data analysis for source attribution, guidance for interpreting WGS analyses is lacking. Here, we provide the framework for interpreting WGS analyses used by the Food and Drug Administration’s Center for Food Safety and Applied Nutrition (CFSAN). We based this framework on the experiences of CFSAN investigators, collaborations and interactions with government and industry partners, and evaluation of the published literature. A fundamental question for investigators is whether two or more bacteria arose from the same source of contamination. Analysts often count the numbers of nucleotide differences [single-nucleotide polymorphisms (SNPs)] between two or more genome sequences to measure genetic distances. However, using SNP thresholds alone to assess whether bacteria originated from the same source can be misleading. Bacteria that are isolated from food, environmental, or clinical samples are representatives of bacterial populations. These populations are subject to evolutionary forces that can change genome sequences. Therefore, interpreting WGS analyses of foodborne bacteria requires a more sophisticated approach. Here, we present a framework for interpreting WGS analyses that combines SNP counts with phylogenetic tree topologies and bootstrap support. We also clarify the roles of WGS, epidemiological, traceback, and other evidence in forming the conclusions of investigations. Finally, we present examples that illustrate the application of this framework to real-world situations.

The analysis of next-generation sequence (NGS) data is often a fragmented step-wise process. For example, multiple pieces of software are typically needed to map NGS reads, extract variant sites, and construct a DNA sequence matrix containing only single nucleotide polymorphisms (i.e., a SNP matrix) for a set of individuals. The management and chaining of these software pieces and their outputs can often be a cumbersome and difficult task. Here, we present CFSAN SNP Pipeline, which combines into a single package the mapping of NGS reads to a reference genome with Bowtie2, processing of those mapping (BAM) files using SAMtools, identification of variant sites using VarScan, and production of a SNP matrix using custom Python scripts. We also introduce a Python package (CFSAN SNP Mutator) that when given a reference genome will generate variants of known position against which we validate our pipeline. We created 1,000 simulated Salmonella enterica sp. enterica Serovar Agona genomes at 100× and 20× coverage, each containing 500 SNPs, 20 single-base insertions and 20 single-base deletions. For the 100× dataset, the CFSAN SNP Pipeline recovered 98.9% of the introduced SNPs and had a false positive rate of 1.04 × 10 −6 ; for the 20× dataset 98.8% of SNPs were recovered and the false positive rate was 8.34 × 10 −7 . Based on these results, CFSAN SNP Pipeline is a robust and accurate tool that it is among the first to combine into a single executable the myriad steps required to produce a SNP matrix from NGS data. Such a tool is useful to those working in an applied setting (e.g., food safety traceback investigations) as well as for those interested in evolutionary questions.Subjects Bioinformatics

The IL-17 pathway is an established driver of psoriasis pathogenesis. We examined the detailed molecular and cellular effects of blockade of IL-17 signaling in human psoriatic skin before and following treatment with brodalumab, a competitive inhibitor of the IL-17 Receptor A subunit. Thousands of aberrantly expressed genes in lesional skin normalized within 2 weeks following brodalumab treatment, with conversion of the lesional psoriasis transcriptome to resemble that seen in nonlesional skin. Keratinocyte-expressed genes appeared to normalize rapidly, whereas T cell–specific normalization occurred over six weeks. The three IL-17 ligand genes that are upregulated in lesional skin, IL17A, IL17C, and IL17F, were all downregulated in a dose-dependent manner following brodalumab treatment. Cellular measures also showed a similar pattern with dramatic decreases in keratinocyte hyperplasia within one week, and decreases in infiltrating leukocytes occurred over a longer timescale. Individuals with the highest brodalumab exposure showed normalization of both IL-17–responsive genes and the psoriasis transcriptome, whereas subjects with lower exposures showed transient or incomplete molecular responses. Clinical and molecular response appeared dependent on the extent of brodalumab exposure relative to the expression of IL-17 ligand genes, and reduction of IL-17 signaling into the nonlesional range was strongly correlated with normalization of the psoriasis transcriptome. These data indicate that blockade of IL-17 signaling in psoriatic skin leads to rapid transcriptomal changes initially in keratinocyte-expressed genes, followed by normalization in the leukocyte abnormalities, and demonstrates the essential role of the IL-17R on keratinocytes in driving disease pathogenesis.

BackgroundIn psoriasis, only limited overlap between sets of genes identified as differentially expressed (psoriatic lesional vs. psoriatic non-lesional) was found using statistical and fold-change cut-offs. To provide a framework for utilizing prior psoriasis data sets we sought to understand the consistency of those sets.Methodology/Principal FindingsMicroarray expression profiling and qRT-PCR were used to characterize gene expression in PP and PN skin from psoriasis patients. cDNA (three new data sets) and cRNA hybridization (four existing data sets) data were compared using a common analysis pipeline. Agreement between data sets was assessed using varying qualitative and quantitative cut-offs to generate a DEG list in a source data set and then using other data sets to validate the list. Concordance increased from 67% across all probe sets to over 99% across more than 10,000 probe sets when statistical filters were employed. The fold-change behavior of individual genes tended to be consistent across the multiple data sets. We found that genes with <2-fold change values were quantitatively reproducible between pairs of data-sets. In a subset of transcripts with a role in inflammation changes detected by microarray were confirmed by qRT-PCR with high concordance. For transcripts with both PN and PP levels within the microarray dynamic range, microarray and qRT-PCR were quantitatively reproducible, including minimal fold-changes in IL13, TNFSF11, and TNFRSF11B and genes with >10-fold changes in either direction such as CHRM3, IL12B and IFNG.Conclusions/SignificanceGene expression changes in psoriatic lesions were consistent across different studies, despite differences in patient selection, sample handling, and microarray platforms but between-study comparisons showed stronger agreement within than between platforms. We could use cut-offs as low as log10(ratio) = 0.1 (fold-change = 1.26), generating larger gene lists that validate on independent data sets. The reproducibility of PP signatures across data sets suggests that different sample sets can be productively compared.