Data were analyzed using the Illumina TSO500 Local app software version 2.0.1.4. As quality filtering criteria, we set median insert size as 75 or more and percentage of target bases with coverage greater than 100× as 75 or more. Then, we annotated Variant Call Format (vcf) files with Annovar,26 (link) applying the following filters: (1) quality, 20 or higher; (2) read depth, 100× or more; (3) variant read frequency, 5% or greater; (4) exclusion of nonexonic, nonsplicing, and synonymous variants; and (5) frequency in population, 1% or less. After that, we annotated variants with Varsome, classifying them as benign, pathogenic, or of uncertain significance.27 (link)We extracted only pathogenic variants and a list of genes with at least one variant. We separated samples in two ways: the first one according to Crohn and non-Crohn disease and the second one according to SRC type or non-SRC type (including PCC-NOS and combined poorly cohesive NOS and SRC carcinomas). Then, we generated oncoprints (ComplexHeatmap R package)28 (link) with all the samples.
Categorical and continuous data were summarized as percentages and medians, and comparisons between groups were performed using the Fisher exact test and Mann-Whitney U test, respectively. A P value less than .05 was considered statistically significant.
Categorical and continuous data were summarized as percentages and medians, and comparisons between groups were performed using the Fisher exact test and Mann-Whitney U test, respectively. A P value less than .05 was considered statistically significant.