Biology subjects

Gallopin, M.

Publications and source records attributed to Gallopin, M..

10 recordsLinked to original sources

bioRxiv2026

phyloDS: phylogeny-aware differential splicing across species, beyond reference annotations

Alternative splicing diverges rapidly between species, yet its comparative analysis lags far behind that of gene expression: junction orthology is defined on a single reference genome, species are treated as independent samples, and detected shifts are rarely grounded in sequence-level evidence. We present phyloDS, an RNA-seq--driven framework that resolves all three limitations. phyloDS defines orthologous splice junctions on multiple sequence alignments of one-to-one orthologous genes, tests their differential usage with a regularized Ornstein--Uhlenbeck model, and cross-validates detected shifts against splice-site predictions made from genomic sequence alone. Applied to primate brain and rodent heart transcriptomes, phyloDS recovered up to 2.9-fold more shared orthologous splice junctions than a reference-based strategy and kept type~I error near nominal where competing methods inflate it more than two-fold. It identified 234 and 502 differentially used orthologous splice junctions in hominoids and mole-rats (86 and 141 unannotated), revealing a mitochondrial and contractile splicing signature in mole-rats.

bioinformatics↗

bioRxiv2026

Interspecies Differential Gene Expression Analysis with Regularized Phylogenetic Linear Models

Comparative transcriptomic datasets are increasingly used to investigate the molecular basis of phenotypic diversification across species. However, finding genes that are differentially expressed (DE) between lineages remains challenging, for two main reasons. First, the random evolutionary drift can blur the signal left by lineage-specific shifts in mean expression, and induces phylogenetic correlations that, if ignored, can widely inflate the False Discovery Rate (FDR), i.e., the amount of spuriously detected genes. Second, DE analysis from RNA-Seq data involves multiple testing on many genes for a small number of individual measurements with high noise, and requires dedicated statistical tools. Traditional DE tools, such as limma, and classical Phylogenetic Comparative Methods (PCMs), such as the Expression Variance and Evolution (EVE) model, are both designed to tackle one of these two challenges alone, but both fail in the context of inter-species RNA-Seq data. In this work, we present phyloDE, a new tool for inter-species DE, that aims at taking the best from both approaches. On simulations based on a recently published four-species rodent dataset, we show that, contrary to other methods, phyloDE correctly controls the FDR in all settings, while keeping a reasonable power. When reanalyzing the empirical dataset, phyloDE discovers more DE genes that exhibit consistent changes in their cis-regulatory landscape compared to EVE in all the experimental settings. The method is implemented in R, with an interface inheriting from limma.

evolutionary biology↗

bioRxiv2025

Metappuccino: Large Language Model-driven Reconstruction of Sequence Read Archive Metadata for Cancer Research

MotivationHigh-throughput RNA-sequencing has significantly advanced transcriptomic profiling in on-cology. Millions of RNA-seq datasets have accumulated in public databases such as the Sequence Read Archive-SRA. However, fragmented, ambiguous or missing metadata can severely limit accurate cohort selection, introduce bias and delay discoveries. ResultsTo address these issues, we introduce Metappuccino : a metadata enrichment tool based on a fine-tuned Mistral-7B-Instruct large language model with low-rank-adaptation (LoRA). Metappuccino can extract or infer 19 key metadata classes (e.g. organ, disease, cell type) from unstructured text. Fine-tuning was conducted with careful partitioning and training design to preserve the models generalisation capacity, reduce data leakage, and ensure robust, context-aware inference rather than memorisation. When possible, the inferred outputs are mapped to standardised ontologies, such as Cellosaurus, Disease Ontology and Uberon, to produce consistent metadata. As a result, the fine-tuned model achieves significantly improved class prediction accuracy over the base model, performing at least as well as recent large open-source models. Furthermore, it reduces inference time by up to at least two compared to the baseline models. As a pipeline, Metappuccino complements the LLM with well-established Natural Language Processing techniques from the literature to further improve performance. By enriching the metadata of under-annotated sequences, Metappuccino creates greater value from public RNA-seq datasets, with potential applications extending beyond oncology transcriptomics. Availability and ImplementationThe source code of Metappuccino is available on GitHub : github. com/chumphati/Metappuccino. The fine-tuned LLM, MetappuccinoLLModel, is available on Hugging Face : huggingface.co/chumphati/MetappuccinoLLModel. Both repositories are released under Apache-2.0 license. Contactfiona.hak@i2bc.paris-saclay.fr, daniel.gautheret@universite-paris-saclay.fr, melina.gallopin@i2bc.paris-saclay.fr

bioinformatics↗

bioRxiv2024

Comparing machine learning models for predicting mutation status in Acute Myeloid Leukemia patients using RNA-seq data

Acute Myeloid Leukemia (AML) is a highly heterogeneous disease. The current AML classifications are based mainly on molecular markers, including cytogenetics features, fusion genes, and the presence or absence of mutations. In this study, we investigated mutation status in AML patients through RNA-seq data in link with differential gene expression. We applied seven machine learning algorithms to identify the presence or absence of NPM1, IDH1/IDH2, and FLT3-ITD mutations, reaching 95%, 93%, and 87% accuracy, respectively. In each case, the best performing models were complex models, suggesting highly complex biological processes at work behind AML.

bioinformatics↗

bioRxiv2024

KaMRaT: a C++ toolkit for k-mer count matrix dimension reduction

SummaryKaMRaT is a program for processing large k-mer count tables extracted from high throughput sequencing data. Major functions include scoring k-mers based on count statistics, merging overlapping k-mers into longer contigs and selecting k-mers based on their presence in certain samples. KaMRaT s main application is the reference-free analysis of multi-sample and multi-condition datasets from RNA-seq, as well as ChiP-seq or ribo-seq experiments. KaMRaT enables the identification of condition-specific or differential sequences, irrespective of any gene or transcript annotation. Implementation and availabilityKaMRaT is implemented in C++. Source code and documentation are available via https://github.com/Transipedia/KaMRaT. Container images are available via https://hub.docker.com/r/xuehl/kamrat.

bioinformatics↗

bioRxiv2023

Visual integration of omics data to improve 3D models of fungal chromosomes

The functions of eukaryotic chromosomes and their spatial architecture in the nucleus are reciprocally dependent. Hi-C experiments are routinely used to study chromosome 3D organization by probing chromatin interactions. Standard representation of the data has relied on contact maps that show the frequency of interactions between parts of the genome. In parallel, it has become easier to build 3D models of the entire genome based on the same Hi-C data, and thus benefit from the methodology and visualization tools developed for structural biology. 3D modeling of entire genomes leverages the understanding of their spatial organization. However, this opportunity for original and insightful modeling is under exploited. In this paper, we show how seeing the spatial organization of chromosomes can bring new perspectives to Hi-C data analysis. We assembled state-of-the-art tools into a workflow that goes from Hi-C raw data to fully annotated 3D models and we re-analysed public Hi-C datasets available for three fungal species. Besides the well-described properties of the spatial organization of their chromosomes (Rabl conformation, hypercoiling and chromosome territories), our 3D models highlighted i) in Saccharomyces cerevisiae, the backbones of the cohesin anchor regions, which were aligned all along the chromosomes, ii) in Schizosaccharomyces pombe, the oscillations of the coiling of chromosome arms throughout the cell cycle and iii) in Neurospora crassa, the massive relocalization of histone marks in mutants of heterochromatin regulators. 3D modeling of the chromosomes brings new opportunities for visual integration. This holistic perspective supports intuition and lays the foundation for building new concepts.

genomics↗

bioRxiv2022

Detection of orthologous genes with expression shifts linked to nickel hyperaccumulation across Eudicots

The remarkable capacity of plants to tolerate and accumulate tremendous amount of nickel is a complex adaptative trait that appeared independently in more than 700 species distributed in about fifty families. Nickel hyperaccumulation is thus proposed as a model to investigate the evolution of complex traits in plants. However, the mechanisms involved in nickel hyperaccumulation are still poorly understood in part because comparative transcriptomic analyses struggle to identify genes linked to this trait from a wide diversity of species. In this work, we have implemented a methodology based on the quantification of the expression of orthologous groups and phylogenetic comparative methods to identify genes which expression is correlated to the nickel hyperaccumulation trait. More precisely, we performed de novo transcriptome assembly and reads quantification for each species on its own transcriptome using available RNA-Seq datasets from 15 nickel hyperaccumulator and non-accumulator species. Assembled contigs were associated to orthologous groups built using proteomes predicted from completed plant genome sequences. We then analyzed the transcription profiles of 5953 orthologous groups from distant species using a phylogenetic ANOVA. We identified 31 orthologous groups with an expression shift associated with nickel hyperaccumulation. These orthologous groups correspond to genes that have been previously implicated in nickel accumulation, and to new candidates involved in this trait. We thus believe that this method can be successfully applied to identify genes linked to other complex traits from a wide diversity of species.

plant biology↗

bioRxiv2022

Benchmark of Differential Gene Expression Analysis Methods for Inter-species RNA-Seq Data using a Phylogenetic Simulation Framework

Inter-species RNA-Seq datasets are increasingly common, and have the potential to answer new questions on gene expression patterns across the evolution. Single species differential expression analysis is a now well studied problem, that benefits from sound statistical methods. Extensive reviews on biological or synthetic datasets have provided the community with a clear picture on the relative performances of the available tools in various settings. Such benchmarks are still missing in the inter-species gene expression context. In this work, we take a first step in this direction by developing and implementing a new simulation framework. This tool builds on both the RNA-Seq and the Phylogenetic Comparative Methods literatures to generate realistic count datasets, while taking into account the phylogenetic relationships between the samples. We illustrate the features of this new framework through a targeted simulation study, that reveals some of the strengths and weaknesses of both the classical and phylogenetic approaches for inter-species differential expression analysis. The tool has been integrated in the R package compcodeR freely available on Bioconductor.

bioinformatics↗

bioRxiv2021

The contribution of uncharted RNA sequences to tumor identity in lung adenocarcinoma

BackgroundTranscriptome analysis of cancer tissues has been instrumental in defining tumor subtypes, diagnostic signatures and cancer regulatory networks. Cancer transcriptomes are still predominantly analyzed at the level of gene expression. Few studies have addressed transcript-level variations, and most of these only looked at splice variants. Previously we introduced a k-mer based, reference-free method, DE-kupl, that performs differential analysis of RNA-seq data at the k-mer level, which enables distinguishing RNAs differing by a single nucleotide. Here we evaluate the significance of differential events discovered by this method in two independent lung adenocarcinoma RNA-seq datasets (N=583 and N=154). ResultsFocusing on differential events in a tumor vs normal setting, we found events in endogenous repeats, alternative splicing and polyadenylation sites, long non-coding RNAs, retained introns and unmapped RNAs. Replicability was highly significant for most event classes (assessed by comparing to events shared between unrelated tumors). Overall about 160,000 differential k-mer contigs were shared between datasets, including a large set of sequences from hypervariable genes such as immunoglobulins, SFTP and mucin genes. Most interestingly, we identified a set of novel tumor-specific long non-coding RNAs in intergenic and intronic regions. We found that expressed endogenous transposons defined two major groups of patients (high/low repeat expression) with distinct clinical characteristic. A number of repeats, intronic RNAs and lincRNA achieved strong patient stratification in univariate or multivariate survival models. Finally, using antigen presentation prediction, we identified 55 contigs predicted to produce recurrent tumor-specific antigens. ConclusionsK-mer based RNA-seq analysis enables description of cancer transcriptomes at nucleotide precision, independently of prior transcript annotation. Application to lung cancer data uncovered events stemming from a wide variety of transcriptional and postranscriptional mechanisms. Among those events, a significant subset was replicable between cohorts, thus constituting novel RNA hallmarks of cancer. The code is available at: https://github.com/Transipedia/dekupl-lung-cancer-inter-cohort.

bioinformatics↗

bioRxiv2019

Blind exploration of the unreferenced transcriptome reveals novel RNAs for prostate cancer diagnosis

The broad use of RNA-sequencing technologies held a promise of improved diagnostic tools based on comprehensive transcript sets. However, mining human transcriptome data for disease biomarkers in clinical specimens is restricted by the limited power of conventional reference-based protocols relying on uniquely mapped reads and transcript annotations. Here, we implemented a blind reference-free computational protocol, DE-kupl, to directly infer RNA variations of any origin, including yet unreferenced RNAs, from high coverage total stranded RNA-sequencing datasets of tissue origin. As a bench test, this protocol was powered for detection of RNA subsequences embedded into unannotated putative long noncoding (lnc)RNAs expressed in prostate cancer tissues. Through filtering and visual inspection of 1,179 candidates, we defined 21 lncRNA probes that were further validated for robust tumor-specific expression by NanoString single molecule-based RNA measurements in 144 tissue specimens. Predictive modeling yielded a restricted probe panel enabling over 90% of true positive detection of cancer in an independent dataset from The Cancer Genome Atlas. Remarkably, this clinical signature made of only 9 unannotated lncRNAs largely outperformed PCA3, the only RNA biomarker approved by the Food and Drug Administration agency, specifically, in detection of high-risk prostate tumors. The proposed reference-free computational workflow is modular, highly sensitive and robust and can be applied to any pathology and any clinical application.

genomics↗