Sample information.

(A) Tumour sample information. Identifiers and metadata are listed for DFT1, DFT2, and non-DFTD tumours (non-DFTD refers to tumours which are neither DFT1 nor DFT2). Each row refers to a single tumour, with multiple samples from the same individual devil linked by a common individual ID and a digit following the “T,” e.g., 31T1 and 31T2 are two tumours sampled from the same host. Further information on devils with multiple tumours can be found in S4 Table. In some cases, “T1” and “T2” refer to the same tumour sampled twice, either at the same or different time points; these are labelled as “duplicate DNA extraction from the same tumour” (same time point) or “Time-course” (different time points). Cell line samples are also labelled “T,” and samples from the same cell line collected at the same or at different passages are labelled “T1” and “T2,” with those collected at the same time point labelled “Duplicate cell pellet.” In addition, the following group of cell lines and experimental transmissions all derive from the same parental cell line (87T, 87T2, 90T, 90T2, 114T, 363T1, 993T1.1, 993T1.2, 994T1.1, 995T1.1, 995T1.2, 996T1, 996T2). Cell lines used in the study are also known by the following external identifiers: 85T (Half-Pea); 86T, 86T2 (1426); 87T, 87T2, 90T, 90T2, 114T, 363T1, 993T1.1, 993T1.2, 994T1.1, 995T1.1, 995T1.2, 996T1, 996T2 (C5065); 88T, 88T2 (4906); 204T (Ed); 202T2 (RV, TD467); 203T3 (SN, TD500). “Matched” and “unmatched” refer to availability of host (see (B)). Clade and clade group (S1 Fig) are listed for each tumour, as well as mitochondrial haplotype group and genotype (S1 Fig and S2 Table). Tumour purity estimates from copy number analysis, MIP genotyping, and mitochondrial analysis are shown. Tumour data, including sampling location, latitude, longitude, date of sampling, host age, and sex (M, male; F, female), are provided. “Singleton” and “multiple” denote individuals from which either a single tumour or multiple tumours were sampled as part of the set, and information on tumour body location is provided, when known. Tumours sampled from an internal body location are marked “metastasis.” Information on the diagnosis of non-DFTD tumours is provided. Cell lines and tumour biopsies derived from experimental transmission trials involving either cell lines or tissue biopsies are marked [26], with inoculation date provided, if known. Tumour strain refers to cytogenetic analysis [13]. Subsets of samples included in the phylogenetic tree (Fig 1A and S1 Fig) and in geographical analyses (Fig 1B–1D and S2 Fig) are indicated, including those that are italicised in S1 Fig. Subsets of samples for which different data types are available (mtDNA haplotype, CNVs, and MIPs genotypes) are listed. Samples which are duplicate biopsies sampled from the same tumour at the same time, duplicate cell pellets from a single cell culture, or which are samples from the same tumour or cell line collected on different dates (time course) are indicated. M5 genotype is provided, with blank indicating that M5 is retained. Tetraploid tumours are marked, and the tetraploid lineage to which each tetraploid tumour belongs is listed in parentheses (S4 Fig). CNV burdens and the width of genome affected (bp) are presented, with separate data for all CNVs and clonal CNVs only. CNV strings are concatenated lists of CNV IDs (S3 Table) and “linked CNVs” groups together those CNVs that cooccur in the same direction (gain, loss) in the same samples and which may have occurred as a single event (S3 Table); CNVs present in a sample that do not form part of a linkage group are included in this count as singleton groups. Fig 2C and S5 Fig show data for “linked CNVs,” with CNVs occurring after whole genome duplication counted as half a CNV in order to correct for genome size. Subclonal CNVs could not be called in tetraploid tumours and the “clonal only” fields have been left blank in these tumours. Telomere length data, presented in S12 Fig, are available in column labelled “Telseq_Length_Estimate.” “SangerStudyIDs” contains additional sample identifiers. (B) Normal devil sample information. Sample ID, microchip (if known), trapping location, gender, and year of sampling are indicated. If the individual is a matched host from a tumour in the study it is marked “matched,” otherwise, it is labelled “unmatched.” Availability of WGS and MIP data are marked. Telomere length data, presented in S12 Fig, are available in column labelled “Telseq_Length_Estimate.” “SangerStudyIDs” contains additional sample identifiers. (XLSX)