Portal de Periódicos da CAPES

Abstracts of the 12th European Cytogenomics Conference 2019

2019; BioMed Central; Volume: 12; Issue: S1 Linguagem: Inglês

10.1186/s13039-019-0439-z

1755-8166

Alain Pinton, Nicolas Mary, Harmonie Barasc, Nathalie Bonnet, Anne Calgaro, Pauline Berlib, Stéphane Ferchaud, Isabelle Raymond‐Letron,

Gene expression and cancer classification

Invited Lecture AbstractsL1 Somatic chromosomal mosaicismJoris Robert Vermeesch (Joris.vermeesch@kuleuven.be)Center for Human Genetics, KULeuven, Leuven, BelgiumIt is generally assumed that all cells in the body are diploid. However, single cell genomic analysis is demonstrating that chromosomal mosaicism is common and that the mitotic error rate is significantly higher than previously assumed. We have been mapping this chromosomal instability starting from the beginning of life, from the zygote to the elderly. To map chromosomal mosaicism, we have been developing methods to accurately map chromosomal changes as well as methods to map haplotypes in single cells. The methods reconstruct genome-wide haplotype architectures as well as the copy-number and segregational origin of those haplotypes by employing phased parental genotypes and deciphering WGA-distorted SNP B-allele fractions. In another approach, we can map chromosomal anomalies by mapping the DNA fragments present in liquid biopsies and subsequently deconvoluting the signals. By looking at both haplotypes and copy numbers we are acquiring novel insights into chromosomal missegregation during the early cleavages, in placenta and in other somatic tissues. I will provide an overview of distinct and unique segregational aberrations.L2 The future of cytogenomics in the diagnosticsClaudia Haferlach (claudia.haferlach@mll.com)MLL Munich leukemia laboratory, Max-Lebsche-Platz 31, 81377 München, GermanyThe spectrum of genomic abnormalities present in human diseases deriving from germline and/or acquried genetic alterations is wide and encompasses gross and submicroscopic aberrations including copy number alterations, structural variants and small nucleotide variants. Thus, to date for a comprehensive genetic work-up a set of cytogenetic and molecular genetic techniques are performed. As sequencing technologies have evolved rapidly whole exome sequencing (WES), whole genome sequencing (WGS), and whole transcriptome sequencing (WTS) have already reached or are ready to be tested in a routine diagnostic setting. WGS provides the possibility to capture all genomic information in a single assay. However, reliable studies in a diagnostic setting are required to determine whether WGS can replace current techniques.At MLL WGS was performed up to now in more than 4000 samples from patients with various hematological malignancies in order to evaluate the feasibility of WGS in a routine diagnotic setting and the impact WGS might have on the diagnostic work-up of hematological neoplasms in future. First analyses revealed a high detection rate by WGS of genomic abnormalities identified by standard diagnostic procedures. 555 of 574 (96.7%) balanced rearrangements detected by chromosome banding analysis were also identified by WGS. Further WGS detected 60 recurrent balanced rearrangements that were missed by CBA due to cytogenetically cryptic fusions or insufficient in vitro proliferation of the aberratin in vitro. Comparing 18,337,602 positions 18,031,728 (98%) yielded the same result with genomic array analysis and WGS with respect to gain, loss or normal copy number status, respectively.The most challenging part of using WGS as a diagnostic tool is the elimination of sequencing and alignment artefacts as well as the clinical interpretation of rare variants. Thus, we built a pipeline integrating several data bases in order to facilitate data interpretation. The next steps on the road towards a diagnostic tool are the validation of copy number alterations, structural variants, and small nucleotide variants identified in addition to standard diagnostics and the determination of the coverage necessary to detect small clones relevant for patient care.The technological advances will change future diagnostics dramatically challenging geneticists to transform the huge amount of genetic data into improved classification of diseases and individualization of treatment in order to improve patients outcome.L3 Liquid biopsies in patients with cancerMichael R. Speicher (Michael.speicher@medunigraz.at)Medical University of Graz, Institute of Human Genetics, Graz, AustriaPrecision oncology seeks to leverage molecular information about cancer to improve patient outcomes and to this end tissue biopsies are widely used to characterize tumor genomes. Recently, attention has been turning to minimally-invasive liquid biopsies, which enable analysis of tumor components (including circulating tumor DNA [ctDNA]) in bodily fluids such as blood. Analyses of ctDNA has been used to track the evolutionary dynamics and heterogeneity of tumors. Furthermore, ctDNA analyses can detect very early emergence of therapy resistance, residual disease, and recurrence. Our group has developed several methods for the analysis of ctDNA. To date, we have analyzed more than 4,000 plasma samples from patients with cancer (breast, prostate, colon, renal and lung carcinoma), which allowed us to estimate the dynamics of clonal evolution of tumor genomes and to identify mechanisms of resistance against given therapies.Recently we leveraged the fact that plasma DNA is nucleosome protected DNA. After whole-genome sequencing appropriate bioinformatics including support vector machines allowed the mapping of nucleosome positions based on the genomic sequencing coverage of plasma DNA fragments. For example, the genomic sequencing coverage of plasma DNA fragments around transcription start sites (TSSs) has a distinct pattern allowing the identification of actively transcribed genes of cells releasing their DNA into the circulation. The expression levels of genes in the corresponding tumor were reflected by the coverage around the TSSs in plasma of patients with cancer. Another approach, based on similar principles, allows assessment of TF activity based on cell-free DNA sequencing and nucleosome footprint analysis. To this end, we analyzed whole genome sequencing data for >1,000 cell-free DNA samples from cancer patients and healthy controls using a newly developed bioinformatics pipeline that infers accessibility of TF binding sites from cell-free DNA fragmentation patterns. We observed tumor-specific patterns, including accurate prediction of tumor subtypes in prostate cancer, with important clinical implications for the management of patients. Furthermore, we show that cell-free DNA TF profiling is capable of detection of early-stage colorectal carcinomas. The great potential of liquid biopsies makes ctDNA analyses to a promising tool for precision medicine.L4 Chromosome banding: the end of the Dark AgesFelix Mitelman (felix.mitelman@med.lu.se)Department of Clinical Genetics, Institute of Laboratory Medicine, University of Lund, Lund, SwedenThe introduction of the first chromosome banding technique (Q-banding) by Lore Zech and Torbjörn Caspersson 50 years ago completely revolutionized cytogenetic analysis. Whereas formerly identification of chromosomes was restricted to chromosome groups based on size and centromere location, now each chromosome, chromosome arm, and even individual chromosome regions could be precisely identified on the basis of their unique banding pattern. Very soon, several other banding techniques were developed, for example, G-, R-, C-, and NOR-banding, each having its own specific properties and applications.The discovery of chromosome banding had an enormous impact and ushered in an unparalleled period of advancement in cytogenetics as well as in several areas of biology and medicine. The possibility to identify and exactly characterize chromosome aberrations in humans laid the foundation for the clinical application of cytogenetic analyses and established clinical cytogenetics as a medical specialty. Numerous constitutional chromosome abnormality syndromes were delineated and the detection of such aberrations in affected individuals or through prenatal diagnostic procedures soon became standard practice. At the same time, consistent or even specific cancer-associated chromosome changes, not imagined before this era, were disclosed in most tumor types. These findings had important implications. Characteristic acquired chromosome aberrations became an important diagnostic tool and also provided a means to unravel pathogenetic mechanisms by pinpointing the location of cancer-initiating genes.The major contribution of banding techniques in basic research in general was the mapping of genes on chromosomes, refined substantially by the subsequent development of high-resolution banding. The utilization of highly elongated pro-metaphase or prophase chromosomes provided even greater precision than conventional banding by revealing more than twice the number of bands seen at metaphase. The assignment and localization of genes to chromosomes at the 850 sub-band level were central to the construction of genetic maps, and high-resolution banding played an important role in the verification of gene order in such maps. It is difficult to overstate the value of this contribution of chromosome banding to the effort to sequence the human genome.L5 Structural variation in the 3D genomic era: Implications for disease and evolutionDarío G. Lupiáñez (Dario.Lupianez@mdc-berlin.de)Institute for Medical Systems, Biology, Max- Delbrück Center for Molecular, Medicine, Berlin- Buch, Germany3D spatial organization is an inherent property of the vertebrate genome to accommodate the roughly 2m of DNA in the nucleus of a cell. On a larger scale, chromosomes display a nonrandom nuclear organization highly influenced by their gene density and transcriptional status. On a subchromosomal scale, the 3D organization of chromatin brings pairs of genomic sites that lie far apart along the linear genome into proximity. Within such organization, topologically associating domains (TADs) emerge as a fundamental structural unit that guides regulatory elements to their cognate promoters to induce transcription (Lupiáñez et al., 2016).Structural and quantitative chromosomal rearrangements, collectively referred to as structural variation (SV), contribute to a large extent to the genetic diversity of the human genome and thus are of high relevance for cancer genetics, rare diseases and evolutionary genetics. Recent studies have shown that SVs can not only affect gene dosage but also modulate basic mechanisms of gene regulation (Lupiáñez et al., 2015; Franke et al., 2016; Will et al., 2017; Bianco et al., 2018; Kragesteen et al., 2018). SVs can alter the copy number of regulatory elements or modify the 3D genome by disrupting higher-order chromatin organization such as TADs. As a result of these position effects, SVs can influence the expression of genes distant from the SV breakpoints, thereby causing the appearance of certain pathogenic phenotypes or evolutionary traits. Therefore, the impact of SVs on the 3D genome and on gene expression regulation has to be considered when interpreting the phenotypical consequences of these variant types (Spielmann et al., 2018).In this talk, I will show examples at different genomic loci, highlighting the potential of SVs to induce developmental disease by distinct pathomechanisms. Furthermore, I will discuss about the iberian mole Talpa occidentalis, a unique case of true XX mammalian hermaphroditism, and a prominent example of how SVs can also be a force of evolutionary innovation.L6 Genome architecture and diseases: the 16p11.2 paradigmAlexandre Reymond (alexandre.reymond@unil.ch)Center for Integrative Genomics, University of Lausanne, Genopode building, CH-1015 Lausanne, SwitzerlandKeywords: CNV, 16p11.2, chromatin, evolution, GWASCopy number changes in 16p11.2 contribute significantly to neuropsychiatric traits. Besides the 600 kb BP4-BP5 (breakpoint) CNV found in 1% of individuals with autism spectrum disorders and schizophrenia and whose rearrangement causes reciprocal defects in head size and body weight, a second distal 220kb BP2-BP3 CNV is a likewise potent driver of neuropsychiatric, anatomical and metabolic pathologies. These two CNVs-prone regions at 16p11.2 are reciprocally engaged in complex chromatin looping and concomitant expression changes, as well as genetic interaction between genes mapping within both intervals, intimating a functional relationship between genes in these regions that might be relevant to pathomechanism.These recurrent pathogenic deletions and duplications are mediated by a complex set of highly identical and directly oriented segmental duplications. This disease-predisposing architecture results from recent, Homo sapiens-specific duplications (i.e. absent in Neandertal and Denisova) of a segment including the BOLA2 gene, the latest among a series of genomic changes that dramatically restructured the region during hominid evolution. Our results show that BOLA2 participates in iron homeostasis and a lower dosage is associated with anemia. These data highlight a potential adaptive role of the human-specific expansion of BOLA2 in improving iron metabolism.Finally, we combined phenotyping of carriers of rare copy variant at 16p11.2, Mendelian randomization and animal modeling to identify the causative gene in a Genome-wide association studies (GWAS) locus for age at menarche. Our interdisciplinary approach allowed overcoming the GWAS recurrent inability to link a susceptibility locus with causal gene(s).L7 Polymer physics predicts the impact on chromatin 3D structure of disease associated structural variantsMario Nicodemi (mario.nicodemi@unina.it)Dipartimento di Fisica "Ettore Pancini" Università di Napoli Federico II Via Cinthia, 21 - Edificio 6 80126 - Naples - ItalyStructural variants (SVs) are a frequent cause of disease and significantly contribute to the variability of our genome, yet their medical impact is usually hard to predict. Recent technologies, such as Hi-C, have revealed that SVs can alter the 3D architecture of chromosomes inducing ectopic contacts between genes and their regulators, leading to mis-expression in congenital diseases. Analogous effects have been also reported in cancer tissues. I discuss how chromosome 3D conformation and folding mechanisms can be understood in a principled way by use of polymer physic [1]. In particular, in the case study of the EPHA4 [2] and PITX1 [3] loci, I illustrate that the effects of pathogenic structural variants can be predicted in-silico, as validated by Hi-C data generated from mouse limb buds and patient-derived fibroblasts.[1] M M. Barbieri, S.Q. Xie, E. Torlai Triglia, A.M. Chiariello, S. Bianco, I. de Santiago, M.R. Branco, D. Rueda, M. Nicodemi*, A. Pombo*, Active and poised promoter states drive folding of the extended HoxB locus in mouse embryonic stem cells. Nature Struct. Mol. Bio, 24, 515 (2017).[2] S. Bianco, D.G. Lupiáñez, A.M. Chiariello, C. Annunziatella, K. Kraft, R. Schöpflin, L. Wittler, G. Andrey, M. Vingron, A. Pombo, S. Mundlos*, M. Nicodemi*, Polymer physics predicts the effects of structural variants on chromatin architecture, Nature Genetics 50, 662 (2018).[3] B.K. Kragesteen, M. Spielmann, C. Paliou, V. Heinrich, R. Schoepflin, A. Esposito, C. Annunziatella, S. Bianco, A.M. Chiariello, I. Jerković, I. Harabula, P. Guckelberger, M. Pechstein, L. Wittler, W.-L. Chan, M. Franke, D.G. Lupiáñez, K. Kraft, B. Timmermann, M. Vingron, A. Visel, M. Nicodemi*, S. Mundlos* and G. Andrey*, Dynamic 3D Chromatin Architecture Determines Enhancer Specificity and Morphogenetic Identity in Limb Development. Nature Genetics 50, 1463 (2018).L8 CNV and Diseases: An overview in constitutional diagnosticsNicole de Leeuw (Nicole.deLeeuw@radboudumc.nl)Department of Human Genetics, Radboud university medical center, Nijmegen, the NetherlandsKeywords: CNV, array, exome, diseaseCopy Number Variants (CNVs), copy number gains or losses ranging in size from less than 1 kb up to many megabases, are frequently identified as the genetic cause in a growing number of disorders. A CNV can involve a single gene leading to specific, phenotypic consequences limited to a single organ or affecting multiple organs, but a CNV may also affect numerous genes resulting in a syndromic phenotype. Some CNV-related clinical features can already be observed prenatally and are present at birth, whereas others develop or become apparent in the first years of life or at a later age.Although CNVs were first considered to be predominantly involved in neurodevelopmental disorders and congenital anomalies, it is now known that they also play a role in many other disorders, ranging from hearing impairment to late-onset diseases. An illustrative overview will be given on the role of CNVs in human disease, predominantly based on the CNV findings in our diagnostic laboratory from a total of more than 22,000 SNP-based arrays and over 31,000 exomes.The ability to detect both nucleotide variants and CNVs in a single exome sequencing test significantly increases the chance to identify the genetic cause for a patient's clinical phenotype, which can help to better define targeted interventions and improve clinical management.L9 The effect of structural variation in the three-dimensional genomeMalte Spielmann (spielmann@molgen.mpg.de)Max Planck Inst. for Molecular Genetics, Ihnestraße 63-73, 14195 Berlin, GermanyStructural and quantitative chromosomal rearrangements (SVs) contribute to a large extent to the genetic diversity of the human genome and thus are of high relevance for cancer genetics, rare diseases and evolutionary genetics. Recent studies have shown that SVs can not only affect gene dosage but also modulate basic mechanisms of gene regulation. SVs can alter the copy number of regulatory elements or modify the 3D genome by disrupting higher-order chromatin organization such as topologically associating domains. As a result of these position effects, SVs can influence the expression of genes distant from the SV breakpoints, thereby causing disease. The impact of SVs on the 3D genome and on gene expression regulation has to be considered when interpreting the pathogenic potential of these variant types. In my talk I will discuss how SVs can modify the 3D organization of the genome by disrupting chromatin domains. I will also describe the phenotypic consequences of genomic disorders resulting from reshuffling of non- coding enhancer sequences and chromatin domain boundaries with the aim of presenting possible strategies for the medical interpretation of SVs in the 3D genome.L10 The role of duplicated genes in human brain evolution and diseaseAarthi Sekar1,2*, Daniela Soto1,2*, José Uribe-Salazar1,2, Gulhan Kaya1, Ruta Sahasrabudhe3, Megan Y. Dennis1,2 1Genome Center, MIND Institute, and Department of Biochemistry & Molecular Medicine; 2Integrative Genetics and Genomics Graduate Group; 3DNA Technologies Sequencing Core Facility, University of California, Davis, CA 95616 Correspondence: Megan Y. Dennis (mydennis@ucdavis.edu)* These authors contributed equally to this work.The human cortex exhibits dramatic anatomical and cognitive differences from those of closely related primate species. Despite a few potential success stories, the underlying genetic contributors to unique human adaptive traits remain undiscovered. We posit that human-specific segmental duplications (HSDs; genomic regions >1 kbp in size with >98% identity) may be a source of neurological innovation and disease that have remained largely understudied. Two HSD genes, SRGAP2C and ARHGAP11B, have been previously implicated in cortex expansion. Using a human haploid-derived (CHM1) BAC resource, we previously performed Pacific Biosciences long-read sequencing to correct the largest, gene-containing HSDs, fixing over 18.2 Mbp in the current human reference build and identifying over 30 additional HSD gene families. Of these, we honed in on a set of ten duplicate gene families with the propensity to be functional today based on their presence in all of humans tested (thousands) and exhibiting gene expression in adult post-mortem tissues from GTEX. To refine our genes to those important in neurodevelopment, we are employing a multifaceted functional approach using cell lines, zebrafish, and mice. Understanding that genetic variation segregating in modern human populations can also inform on if a gene is functional (e.g., an excess of truncating mutations may indicate loss of function), we are leveraging sequence data of HSD gene paralogs. Unfortunately, only 1.7% of HSDs are accessible for variant calling using whole-genome shotgun short-read (Illumina) data from 1000 Genomes Project. Furthermore, 78% of HSD regions are completely depleted for common variants (dbSNP). As such, we are performing targeted long-read sequencing in diverse human populations to accurately detect variants in these typically inaccessible regions. Though a work in progress, if successful, the results of these studies will offer important insights into if/how HSD genes contribute to innovative neurological features that distinguish modern humans from related great ape species.L11 Inversion variants in the human genomeFrancesca Antonacci (francesca.antonacci@uniba.it)Dept. Biology, University of Bari, Bari, ItalyStructural variation is increasingly acknowledged as an important source of human genetic variation accounting for disease and population diversity. Significant advances have been made over the past few years in mapping and characterizing structural variation in the human genome. Inversion polymorphisms, however, represent a relatively unexplored form of structural variation. Although they are not usually associated with alterations in gene copy number and, thus, do not have a primary effect on phenotype, several of the polymorphic inversions identified to date confer a predisposition to further chromosomal rearrangements in subsequent generations. The majority of inversions described in the human genome are flanked by highly identical segmental duplications causing assembly errors in genome references as well as problems for inversion discovery using next-generation sequencing approaches. Combining molecular cytogenetics, genomic approaches, and sequencing of long molecules we recently characterized some of the largest inversion polymorphisms in the human population. We investigated their worldwide population characteristics, established their association to human disease, and unveiled their evolutionary history. Our data shows that inversion polymorphisms are common and some show striking population stratification. Inversions associate with regions predisposed to disease- causing microdeletions and reoccur at a high frequency due to the presence of duplicated sequences at their boundaries. These structural polymorphisms occur at varying frequencies in populations leading to different susceptibility and ethnic predilection.L12 Male infertility in humans, interest of whole exome sequencingPierre F. Ray (PRay@chu-grenoble.fr)University Grenoble Alpes, INSERM U1209, CNRS UMR 5309, Institute for Advanced Biosciences, Team Genetics Epigenetics and Therapies of Infertility, 38000 Grenoble, FranceKeywords: Infertility, spermatogenesis, genetic diagnosis, whole exome sequencingInfertility is currently considered by the World Health Organization (WHO) as a major health concern affecting more than 50 million couples worldwide. An abnormality of the spermogram is found in half of the cases indicating a balanced responsibility between the man and the woman in the occurrence of couple infertility. Male infertility is characterized by a multifactorial etiology often associated with a strong genetic component, especially when it comes to severe phenotypes such as azoospermia and monomorphic teratozoospermia. The most commonly identified genetic causes of male infertility concern chromosomal defects affecting mainly the gonosomes like Klinefelter syndrome or Y microdeletions but also the autosomes as balanced translocations (robertsonian and reciprocal) are often associated with infertility. However, the realization of a karyotype and the screening for microdeletions permit to reach a diagnosis for no more than 20% of men with non-obstructive azoospermia and diagnosis efficiency is much reduced for the milder sperm defects. Hundreds of genes are specifically expressed in the testes and are necessary for spermatogenesis, the occurrence of genetic defects in any of these genes is therefore likely to result in male infertility. The profusion of these genes makes the identification of mutations responsible for male infertility difficult and complex. High throughput sequencing, and in particular whole exome sequencing (WES), is however revolutionizing this field and has recently permitted to identify numerous genes associated with different phenotypes of male infertility.Our Grenoble team "epigenetic genetics and infertility therapies" has been working for 10 years to identify and characterize the causes of male infertility. Using homozygosity mapping techniques we could identify several major genes involved in different male infertility syndromes such as AURKC in macrozoospermia, DPY19L2 in globozoospermia, and DNAH1 in the phenotype of multiple flagellar morphological abnormalities (MMAF). Whole exome sequencing has recently allowed us to be much more efficient and to identify several other genes involved in the MMAF phenotype (CFAP43, CFAP44, CFAP69, AK7, FSIP2, WDR66, ARMC2) or azoospermia (SPINK2) or female infertility (PATL2). Genes localized on the X chromosome have also been identified with TEX11 and ADGRG2 resulting respectively in non-obstructive and obstructive azoospermia.Some small genomic rearrangements (CNVs) have been identified and characterized and are the most frequent defects identified in DPY19L2 and WDR66. These abnormalities can be detected effectively by WES. The use of high throughput sequencing is transforming the diagnosis of male infertility and this technique is expected to become an integral part of the routine diagnosis proposed for the management of infertile.L13 Aneuploidy in humans: new insights into an age-old problemTerry Hassold (terryhassold@wsu.edu)School of Molecular Biosciences, Washington State University, 99164 Pullman WA USAKeywords: aneuploidy, meiotic recombinationAneuploidy is the most common genetic complication of pregnancy, with approximately 0.2-0.3% of newborn infants being trisomic. However, this represents just the tip of a large iceberg, because most aneuploid conceptions die in utero. Indeed, studies of preimplantation embryos suggest that a large proportion, if not a majority, of fertilized human eggs have extra or missing chromosomes. Because the vast majority of errors result from the fertilization of a chromosomally abnormal egg by a normal sperm, attention has focused on why human female meiosis is so error-prone.In this presentation, we will briefly summarize our work indicating that there are multiple routes to female-derived aneuploidy; e.g., studies of model organisms indicating the contribution of errors occurring during the long meiotic arrest stage or as part of the meiotic cell cycle checkpoint machinery. We will also discuss our recent studies, which have focused on analyzing human meiosis “as it happens” in fetal oocytes and in spermatocytes. These studies demonstrate remarkable differences between human males and females in the way in which chromosomes find and synapse with one another, in the packaging of chromatin, and in the control of the meiotic recombination pathway. Further, they indicate that errors in fetal oogenesis - especially those that lead to failure to recombine or abnormally located crossovers - are surprisingly common in humans. Indeed, our observations suggest that the propensity to nondisjoin may be established – at least in part – at the very beginning of the development of the human oocyte.L14 Clonal evolution among different sarcoma subtypesFredrik Mertens (fredrik.mertens@med.lu.se)Division of Clinical Genetics, Department of Laboratory Medicine, Lund University, SE-221 84 Lund, Sweden and Department of Clinical Genetics and Pathology, University and Regional Laboratories Region Skåne, SE-221 85 Lund, SwedenSarcomas are malignant tumors arising in soft tissues or bone. Although genetic analyses have shown that most morphologic subtypes have unique chromosomal features, the pathogenetic mechanisms behind sarcoma development may be broadly categorized into three distinct subgroups; the largest is characterized by complex combinations of chromosomal gains and losses, followed by sarcomas driven by gene fusions or supernumerary ring chromosomes. To compare patterns of clonal evolution in sarcomas arising through these three different mechanisms, we selected sarcomas with complex genomes (myxofibrosarcomas, MFS), gene fusion-driven myxoid liposarcomas (MLS), and amplicon-driven well-differentiated liposarcomas (WDLS) from which we had access to multiple samples during tumor progression; a further requisite was that at least one year should have elapsed between first and last sampling. We also studied multiple samples from some of the primary lesions, in order to evaluate intra-lesional heterogeneity. Clonal heterogeneity was assessed through a combination of chromosome banding, single nucleotide polymorphism (SNP) array, and whole-exome sequencing analyses.We could show that the type of clonal evolution – i.e., whether nucleotide or chromosome level mutations predominate – and the rate by which new mutations accrue vary considerably among the three sarcoma types. In MFS, tumor progression was usually accompanied by accumulation of both chromosome and nucleotide level aberrations. Primary MLS display little intratumoral heterogeneity and few new mutations are found in local recurrences or metastases. WDLS, on the other hand showed extensive inter-cellular variation in terms of chromosome level aberrations; this variation, however, had only minor impact on