Biology subjects

Faure, R.

Publications and source records attributed to Faure, R..

8 recordsLinked to original sources

bioRxiv2026

Sketching the SRA: experiment-wide similarity search across 4.8 million metagenomes

Recent efforts have demonstrated that systematic analysis of the millions of metagenomes in the Sequence Read Archive (SRA) enables groundbreaking biological discoveries. However, SRA-scale comparison of whole metagenomes has remained elusive. Here, we introduce the most comprehensive resource to date for experiment-to-experiment analysis of all SRA metagenomes, comprising FracMinHash sketches (470GB), hypergen sketches (19GB), and an all-pairs Jaccard similarity matrix (90GB), which we release online at https://github.com/RolandFaure/SRA-metagenome-sketches-databases. These resources enable a wide range of analyses, which we demonstrate with several applications. First, we use the FracMinHash sketches to uncover temporal trends, finding that the amount of newly discovered microbial sequence continues to grow, rather than saturate or shrink. Second, we use the hypergen sketches, which enable nearest-neighbor search, to identify co-assembly candidates and thereby improve assembly quality. Third, we use the similarity matrix to analyze all SRA metagenomes, identifying duplicated submissions and over 50K candidates for metadata correction. Together, our resources unlock systematic analysis of metagenomic data at unprecedented scale.

bioinformatics↗

bioRxiv2026

DIANA: Deep Learning Identification and Assessment of Ancient DNA

The field of ancient metagenomics provides insights into past microbiomes, but with a growing dataset size, methods that rely on reference databases have limited scope. Here, we introduce DIANA, a multi-task neural network that predicts key metadata categories from unitig abundances. Trained on 2,597 run accessions (1.72 Tbp of assembled unitig sequences), DIANA accurately identifies sample host (94.6%), community type (90.0%), and material (88.9%) on held-out test data and demonstrates robust generalisation on an independent validation set. A key innovation is DIANAs ability to perform semantic generalisation, correctly classifying samples with labels unseen during training -- such as novel subspecies -- to their appropriate parent categories. By leveraging both known and uncharacterized genomic sequences, DIANA provides a rapid, data-driven system for metadata validation and quality control, accelerating discovery in ancient metagenomics research.

bioinformatics↗

bioRxiv2025

SNooPy: a statistical framework for long-read metagenomic variant calling

Current long-read single-nucleotide variant callers were designed primarily for genomic data--particularly human genomes. While some have been used on metagenomic data, their underlying assumptions and training procedures fail to account for the inherent complexity of metagenomic samples. To date, no long-read variant caller has been purpose-built for metagenomic applications. To address this gap, we present SNooPy, a SNP-calling tool that implements a new statistical framework tailored to long-read metagenomic data. Unlike previous genomic methods, our approach makes no assumptions about the number of haplotypes present, their evolutionary relationships, or their sequence divergence. We demonstrate that SNooPy outperforms both traditional statistical and deep learning-based SNP callers. Our results suggest that future integration of this framework with deep learning approaches could further enhance variant calling performance. SNooPy is freely available on github.com/rolandfaure/snoopy.

bioinformatics↗

bioRxiv2025

Alice: fast and haplotype-aware assembly of high-fidelity reads based on MSR sketching

BackgroundLong-read metagenomic assembly is becoming a critical bottleneck in microbiome analysis, as deep sequencing generates massive datasets that existing methods struggle to assemble while maintaining strain resolution. ResultsWe present Alice, a lightweight long-read assembler that achieves orders-of-magnitude speedups through a new sequence sketching technique, MSR sketching, compatible with classical assembly methods. Alice assembles a 235 Gbp soil metagenome in 5 hours using only 84 GB RAM--a task that causes most competing methods to exhaust our computational resources (500 GB RAM and 7 days runtime). Across diverse benchmarks, Alice delivered strain-resolved assemblies an order of magnitude faster than state-of-the-art approaches, while producing the most complete assemblies in some cases. ConclusionsMSR sketching overcomes computational barriers in metagenomic assembly, enabling fast, memory-efficient strain-resolved analysis of massive datasets. While Alices assemblies were more fragmented than with other assemblers, this approach establishes a promising paradigm for scalable metagenomic analysis.

bioinformatics↗

bioRxiv2024

First chromosome-level genome assembly of the colonial tunicate Botryllus schlosseri

Botryllus schlosseri (Tunicata) is a colonial chordate that has long been studied for its multiple developmental pathways and regenerative abilities and its genetically determined allorecognition system based on a polymorphic locus that controls chimerism and cell parasitism. We present the first chromosome-level genome assembly from an isogenic colony of B. schlosseri clade A1 using a mix of long and short reads scaf-folded using Hi-C. This haploid assembly spans 533 Mb, of which 96% are found in 16 chromosome-scale scaffolds. With a BUSCO completeness of 91.2%, this complete and contiguous B. schlosseri genome assembly provides a valuable genomic resource for the scientific community and lays the foundation for future investigations into the molecular mechanisms underlying coloniality, regeneration, histocompatibility, and the immune system in tunicates.

genomics↗

bioRxiv2024

HairSplitter: haplotype assembly from long, noisy reads

MotivationLong-read assemblers face challenges in discerning closely related viral or bacterial strains, often collapsing similar strains into a single sequence. This limitation has been hampering metagenome analysis, as diverse strains may harbor crucial functional distinctions. ResultsWe introduce a novel software, HairSplitter, designed to retrieve strains from a partially or totally collapsed assembly and long reads. The method uses a custom variant-calling process to operate with erroneous long reads and introduces a new read binning algorithm to recover an a priori unknown number of strains. On noisy long reads, HairSplitter recovers more strains while being faster than state-of-the-art tools, both in the cases of viruses and bacteria. AvailabilityHairSplitter is freely available on GitHub at github.com/RolandFaure/HairSplitter. Contactroland.faure@irisa.fr

bioinformatics↗

bioRxiv2021

GraphUnzip: unzipping assembly graphs with long reads and Hi-C

Long reads and Hi-C have revolutionized the field of genome assembly as they have made highly continuous assemblies accessible for challenging genomes. As haploid chromosome-level assemblies are now commonly achieved for all types of organisms, phasing assemblies has become the new frontier for genome reconstruction. Several tools have already been released using long reads and/or Hi-C to phase assemblies, but they all start from a linear sequence, and are ill-suited for non-model organisms with high levels of heterozygosity. We present GraphUnzip, a fast, memory-efficient and accurate tool to unzip assembly graphs into their constituent haplotypes using long reads and/or Hi-C data. As GraphUnzip only connects sequences in the assembly graph that already had a potential link based on overlaps, it yields high-quality gap-less supercontigs. To demonstrate the efficiency of GraphUnzip, we tested it on a simulated diploid Escherichia coli genome, and on two real datasets for the genomes of the rotifer Adineta vaga and the potato Solanum tuberosum. In all cases, GraphUnzip yielded highly continuous phased assemblies.

bioinformatics↗

bioRxiv2018

A type 2 diabetes disease module with a high collective influence for Cdk2 and PTPLAD1 is localized in endosomes

Despite the identification of many susceptibility genes our knowledge of the underlying mechanisms responsible for complex disease remains limited. Here, we identified a type 2 diabetes disease module in endosomes, and validate it for functional relevance on selected nodes. Using hepatic Golgi/endosomes fractions, we established a proteome of insulin receptor-containing endosomes that allowed the study of physical protein interaction networks on a type 2 diabetes background. The resulting collated network is formed by 313 nodes and 1147 edges with a topology organized around a few major hubs with Cdk2 displaying the highest collective influence. Overall, 88% of the nodes are associated with the type 2 diabetes genetic risk, including 101 new candidates. The Type 2 diabetes module is enriched with cytoskeleton and luminal acidification -dependent processes that are shared with secretion-related mechanisms. We identified new signaling pathways driven by Cdk2 and PTPLAD1 whose expression regulate the association of the insulin receptor with TUBA, TUBB, the actin component ACTB and the endosomal sorting markers Rab5c and Rab11a. Therefore, the interactome of internalized insulin receptors reveals the presence of a type 2 diabetes disease module enriched in new layers of feedback loops required for insulin signaling, clearance and islet biology.\n\nAuthor SummaryAccording to the local hypothesis each complex disease can be linked to a well-defined network called the disease module. A disease module can be defined by the topological properties of protein interaction networks. Given the complexity of the whole interaction map the existence of such disease modules remains largely to be tested. Here, we found a type 2 diabetes disease module in insulin receptor-containing endosomes. The disease module contains new pathways that are associated with both insulin signaling, clearance and secretion. Its co-functionality with islets biology may provide a mechanistic rationale for the exploration of personalized medicine and elaborate new drugs.

genomics↗