Search

SummaryWith the recent rapid advancements in sequencing technology, the analysis of omics data has become increasingly important for biomedical research. However, many laboratories lack the time or expertise to thoroughly analyze such data. Among omics assays, ATAC-Seq (Assay for transposase-accessiblechromatin usingsequencing) and mRNA-Seq (messengerRNA sequencing) are easy and cost-effective methods that are commonly used to define chromatin accessibility and gene expression levels, respectively. Here we introduce Cactus (Chromatinaccessibility and transcriptomicsunifyingsoftware), a pipeline that streamlines the individual or integrated analysis of ATAC-Seq and mRNA-Seq data. The pipeline is written in Nextflow for efficient scaling, caching, and parallelization, and all tools are packaged in Singularity/Docker containers or conda/Mamba virtual environments for simple installation and high reproducibility. Cactus conducts preprocessing on raw sequencing reads, followed by differential analysis between conditions. Results are split into Differential Analysis Subsets (DASs) based on significance threshold, direction of change, annotated genomic regions, and experiment type. Then, Cactus computes enrichments of entries from internal (i.e., DASs) and external (i.e., ontologies, pathways, DNA binding motifs, ChIP-Seq (Chromatin immunoprecipitationsequencing) binding sites, and chromatin states) databases in DASs and presents results in barplots and customizable heatmaps. The visualization of the results is simplified by the generation of merged PDFs, merged tables, and formatted Excel tables. In conclusion, Cactus can assist researchers in gaining comprehensive insights from chromatin accessibility and gene expression data in a quick, easy, and reproducible manner.Availability and implementationCactus can be installed via Nextflow pull jsalignon/cactus. Source code and documentation are available online (https://github.com/jsalignon/cactus).Contactjerome.salignon@ki.seorchristian.riedel@ki.se

Organisms live in dynamic environments. However, most experimental approaches study the function and selection of genes in steady environments. Therefore, natural selection acting on fluctuating environments remains poorly understood. The objective of my project was to determine if some genes are especially important for fitness (growth rate) of yeast cells in oscillating environments. A genomic screen, based on an automation of micro-cultures and on a multiplexing of sequencing libraries, allowed me to measure fitness of thousands of null mutants in periodic stress conditions. I found that predictability of fitness in periodic stress, from fitness in steady environments, varies depending on the specific genes and conditions considered. This way, some mutants have similar growth in steady conditions, and different growth in dynamic conditions. Curiously, some genes play a bivalent role: they strongly favor growth during slow fluctuations, and reduce it during fast fluctuations. I also observed many mutants with higher growth than expected at the highest frequencies of fluctuations. This effect can be partially explained by a loss of environmental sensitivity of those mutants, that continue to divide quickly despite the presence of a stress. Those results show how natural selection can act on mutations in fluctuating environments. They open the door to mechanistic studies of the predictability of fitness in periodic environments.; Les organismes vivent dans des environnements dynamiques. Or la plupart des approches expérimentales étudient la fonction et la sélection des gènes dans des environnements statiques. De ce fait, la sélection naturelle agissant en environnements fluctuants reste mal comprise. L´objectif de mon projet a été de déterminer si certains gènes sont particulièrement importants pour la fitness (taux de croissance) de cellules de levures en environnements oscillants. Un crible génomique, basé sur une automatisation de micro-cultures et sur un multiplexage de banques de séquençage, m´a permis de mesurer la fitness de milliers de mutants nuls en conditions de stress périodique. J´ai trouvé que la prédictibilité de la fitness en environnements périodiques, à partir de la fitness en environnements statiques, diffère selon les gènes et les conditions. Ainsi, certains mutants présentent des croissances similaires en conditions statiques mais différentes en conditions dynamiques. Curieusement, quelques gènes jouent un rôle bivalent : ils favorisent fortement la croissance lors de fluctuations lentes et ils la défavorisent lors de fluctuations rapides. J´ai également observé de nombreux mutants avec une croissance plus élevée qu´attendue aux fréquences de fluctuations les plus rapides. Cet effet s´explique partiellement par une perte de sensibilité environnementale de ces mutants, qui continuent à se diviser rapidement malgré la présence d´un stress. Ces résultats montrent comment la sélection naturelle agit sur les mutations en environnements fluctuants. Ils ouvrent la porte à des études mécanistiques de la prédictibilité de la fitness en environnements périodiques.