The influence of genomic context on mutation patterns in the human genome inferred from rare variants

Understanding patterns of spontaneous mutations is of fundamental interest in studies of human genome evolution and genetic disease. Here, we used extremely rare variants in humans to model the molecular spectrum of single-nucleotide mutations. Compared to common variants in humans and human-chimpanzee fixed differences (substitutions), rare variants, on average, arose more recently in the human lineage and are less affected by the potentially confounding effects of natural selection, population demographic history, and biased gene conversion. We analyzed variants obtained from a population-based sequencing study of 202 genes in >14,000 individuals. We observed considerable variability in the per-gene mutation rate, which was correlated with local GC content, but not recombination rate. Using >20,000 variants with a derived allele frequency >= 10(-4), we examined the effect of local GC content and recombination rate on individual variant subtypes and performed comparisons with common variants and substitutions. The influence of local GC content on rare variants differed from that on common variants or substitutions, and the differences varied by variant subtype. Furthermore, recombination rate and recombination hotspots have little effect on rare variants of any subtype, yet both have a relatively strong impact on multiple variant subtypes in common variants and substitutions. This observation is consistent with the effect of biased gene conversion or selection-dependent processes. Our results highlight the distinct biases inherent in the initial mutation patterns and subsequent evolutionary processes that affect segregating variants.

KRAB zinc-finger proteins contribute to the evolution of gene regulatory networks

Pierre-Yves Joseph Laurent Helleboid, Michaël Imbeault, Didier Trono

The human genome encodes some 350 Kruppel-associated box (KRAB) domain-containing zinc-finger proteins (KZFPs), the products of a rapidly evolving gene family that has been traced back to early tetrapods(1,2). The function of most KZFPs is unknown, but a few have been demonstrated to repress transposable elements in embryonic stem (ES) cells by recruiting the transcriptional regulator TRIM28 and associated mediators of histone H3 Lys9 trimethylation (H3K9me3)-dependent heterochromatin formation and DNA methylation(3-9). Depletion of TRIM28 in human or mouse ES cells triggers the upregulation of a broad range of transposable elements(4,10,11), and recent data based on a few specific examples have pointed to an arms race between hosts and transposable elements as an important driver of KZFP gene selection(5). Here, to obtain a global view of this phenomenon, we combined phylogenetic and genomic studies to investigate the evolutionary emergence of KZFP genes in vertebrates and to identify their targets in the human genome. First, we unexpectedly reassigned the root of the family to a common ancestor of coelacanths and tetrapods. Second, although we confirmed that the majority of KZFPs bind transposable elements and pinpoint cases of ongoing co-evolution, we found that most of their transposable element targets have lost all transposition potential. Third, by examining the interplay between human KZFPs and other transcriptional modulators, we obtained evidence that KZFPs exploit evolutionarily conserved fragments of transposable elements as regulatory platforms long after the arms race against these genetic invaders has ended. Together, our results demonstrate that KZFPs partner with transposable elements to build a largely species-restricted layer of epigenetic regulation.

Nature Publishing Group2017

Inferring the age and strength of mutations evolving under positive selection

Nicky Louise Ormond

A fundamental goal of population genetics is to determine and quantify the interplay of mutation, natural selection, genetic drift and migration in shaping allelic frequency changes that underpin evolution-ary change. In this dissertation, I present computational, empirical and experimental approaches to address this goal. In the first chapter, I develop an approximate Bayesian computation (ABC) approach that co-estimates the selection strength and age of fixed beneficial mutations in single populations, by integrating a range of existing diversity, site frequency spectrum, haplotype and linkage disequilibrium based summary statistics. This approach is then extended to models of selection on standing variation in order to co-infer the frequency at which positive selection began to act upon the mutation. In the second chapter, I apply this method to an empirical study of convergent adaptation of blanched dorsal phenotypes in two lizard species to the newly formed White Sands system of New Mexico. Estimates of the age of the beneficial mutations underpinning the evolution of cryptic coloration are younger than the related geological shift and support a model of adaptation from de novo mutation. In the third chapter, I analyze the experimental evolution of H1N1 influenza virus populations under a combined protocol of two drugs with different modes of action: oseltamivir and favipiravir. Results indicate a complex interplay of mutation, selection and genetic drift, where selective sweeps around oseltamivir resistance mutations hitchhike deleterious mutations owing to the mutagenic effect of favipiravir to fixation. This effect reduces viral fitness and ac-celerates extinction via Mullerâs ratchet, but at the risk of spreading both established and newly emerging oseltamivir resistance mutations.

EPFL2017

Cell lineage tracing in human tissues using somatic mitochondrial mutations

Paul Refinetti

Lineage tracing is a basic concept in developmental biology, as well as diseases related to organ development. A reliable method to track cell lineages in human tissues has the potential to greatly contribute to the fields of oncogenesis, tissue renewal, ageing as well as wound healing. Inherited mitochondrial mutations have been a successful tool in the understanding of human lineage and migrations throughout history. Similarly, somatic mitochondrial mutations have been used to study the patters of cell division and renewal in some human tissues. Each cell has many copies of the mitochondrial genome, a mutation can therefore be present in any fraction of the total number of mitochondrial DNA copies. To study somatic mitochondrial mutations therefore requires mutations to detected, and the mutant fraction to be quantified. This added requirement makes the use of most sequencing techniques very difficult. The first step in the development of lineage tracing methods using somatic mtDNA mutations as a marker, as a marker, is a quantitative, fast and high throughput method of detection. A two step PCR was designed to amplify first the whole mtDNA in large fragments without amplifying nuclear DNA. Second, a set of small fragments that can be scanned for mutations using cycling temperature capillary electrophoresis. In the end, 76% of the mitochondrial genome could be scanned for mutations. Test showed that mutations could be detected and accurately quantified at fractions as low as 1%. Furthermore, the method could be applied to analyse micro-anatomical samples taken by laser capture micro dissection. Lineage tracing in human tumors was achieved by scanning the mitochondrial DNA for mutations in samples extracted from macro-anatomical samples taken across surgically removed and freshly frozen tumor pieces. Tumors carrying a mutation at a sufficient fraction were then finely sampled. The first attempt consisted in taking 286 samples across a slice of a Leydig cell tumor, in which two separate mutations were identified at a high enough fraction to justify further investigation. Both mutations strongly correlated suggesting a single lineage. Since no discernible pattern emerged, the results were validated by taking 96 samples on another slice of tissue, slightly deeper, and observing that in similar positions, similar mutant fractions could be reproduced. The natural evolution was to take multiple successive slices of tissue. On each slice 96 samples,arranged in an

8\times 12

grid are then taken. Analysing the results from19 layers gave the three dimensional reconstruction of a tumor lineage in a tissue volume. The results show that the lineage followed is quite diffused through the volume and does not occupy its majority. A three dimensional reconstruction of a cell lineage was then established for 2 more tumors, and 3 metastasis. Under the assumption that a metastasis is seeded by a single tumor cell, the presence of the same mtDNA mutation in both primary tumor and metastasis guarantees that only cells carrying the mutation are tumor-derived. The three dimensional reconstruction of a primary tumor lineage through a metastasis is therefore the reconstruction of the patters of metastasis growth in its host tissue. Observing that most of the metastasis volume is not composed of tumor derived cells suggests that metastatic cells grow by diffusing through their host tissue and conditioning it to assume its abnormal morphology.

EPFL2017

Cell lineage tracing in human tissues using somatic mitochondrial mutations

Paul Refinetti

8\times 12

EPFL2017