SEARCH

SEARCH BY CITATION

FilenameFormatSizeDescription
path4214-sup-0001-FigureS1.tifTIFF image1502KCopy-neutral loss of heterozygosity in patient RK21. Segmented copy number of the major (red) and minor (blue) alleles in chromosome 3
path4214-sup-0002-FigureS2.tifTIFF image2880KHierarchical clustering of 143 stage III and IV TCGA samples and 48 multi-region samples from eight patients. Heat map columns represent samples and are ordered to match the dendrogram, while rows represent cytobands. Horizontal bars delimit chromosomes. The hierarchical clustering is based on Spearman's rank correlation coefficient between samples. The colour coding indicates the relative copy number of each cytoband in each sample. Significant clusters obtained using 1000 bootstrapping iterations are highlighted by bars below the copy number profiles of the relevant samples, using alternating grey and black for clarity. Multi-region samples are highlighted and reported below the significant clusters bar
path4214-sup-0003-FigureS3.tifTIFF image3064KHierarchical clustering of 440 TCGA samples and 48 multi-region samples from eight patients. Heat map columns represent samples and are ordered to match the dendrogram, while rows represent cytobands. Horizontal bars delimit chromosomes. The hierarchical clustering is based on Euclidean distances between samples. The colour coding indicates the relative copy number of each cytoband in each sample. Significant clusters obtained using 1000 bootstrapping iterations are highlighted by bars below the copy number profiles of the relevant samples, using alternating grey and black for clarity. Multi-region samples are highlighted and reported below the significant clusters bar
path4214-sup-0004-FigureS4.tifTIFF image10269KModel-based clustering of 440 TCGA samples and 48 multi-region samples from eight patients with hmmMix. Cluster profiles are shown on top, green indicates copy number loss and red indicates copy number gain. Clusters and samples are shown below with horizontal white bars separating the clusters, along with delimiting vertical coloured bars on the left. Samples within a cluster are ordered as they were prior to hmmMix computation. Multi-region samples are highlighted by the coloured bar on the left, indicating from which patient they originate
path4214-sup-0005-FigureS5.tifTIFF image4204KNetwork representation of relationships between multi-region and stage III and IV. TCGA samples using Euclidean distances. Samples are represented as nodes (diamonds for stage III samples, circles for stage IV) and Euclidean distances are used to weight the edges connecting them. The network is fully interconnected and all edges are used to create the layout. Only edges corresponding to distances significantly smaller than expected (including the bottom 5% of the unrelated samples distance distribution) are drawn with thick lines. Coloured lines highlight connections between samples from the same tumour; black lines are used for unrelated samples; white node labels indicate regions from untreated tumours; black node labels indicate regions from treated tumours
path4214-sup-0006-FigureS6.tifTIFF image1758KOverall copy-number profiles of all samples used in the hierarchical clustering analysis. The plain black line represents the median copy number of each cytoband across all samples. Vertical bars specify chromosome starts (plain lines) and centromeres (dashed lines). Blue dots indicate cytobands located on a chromosomal arm lost in more than 20% of all TCGA samples and red dots indicate those on arms gained in more than 20% of all TCGA samples
path4214-sup-0007-FigureS7.tifTIFF image906KExpression of genes impacted by recurrent, large copy number aberrations. Normalized gene expression data were obtained from RNA-seq data on the TCGA website, using the 400 samples for which SNP array data were also available. A large aberration was considered present if at least half of the chromosome arm was gained or lost, absent otherwise. The expression of all genes on each analysed chromosome arm was compared between samples presenting the aberration and samples where it was considered absent. Gene expression distribution are displayed with box plots, green for normal samples, red for samples showing a gain (left) and blue for samples showing a loss (right). p values were obtained using a two-tailed t-test
path4214-sup-0008-TableS1.docWord document30KPatient clinical and pathological data
path4214-sup-0009-TableS2.docWord document58KList of the samples most similar to each multi-region profile, as defined by smallest correlation-based and Euclidean distances
path4214-sup-0010-TableS3.docWord document55KNumber of significantly similar samples
path4214-sup-0011-TableS4.docWord document64KNumber of large aberrations (involving at least half of the length of a chromosome arm) per multi-region sample and number of average vs ubiquitous number of aberrations per patient

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.