Biology subjects

Poch, O.

Publications and source records attributed to Poch, O..

8 recordsLinked to original sources

bioRxiv2025

Do 5' regions of human protein-coding genes contain the blueprints for alternative splicing?

To investigate alternative splicing capacity, we statistically compared the properties of human protein-coding genes with multiple transcript isoforms (MISOG) and single transcript isoforms (SISOG). Apart from global exon content, differential features are concentrated in the 5 gene regions, with MISOG presenting complex 5 untranslated region architecture and a distinctive flanking environment around first 5 intron. Importantly, we found that 5 exons are more prone to alternative splicing in MISOG. These results unravel previous observations indicating the importance of 5 gene regions in some transcriptional processes and call for their reassessment in light of the MISOG/SISOG profiles.

bioinformatics↗

bioRxiv2025

TransStop, a genomic language model for the pan-drug prediction of translational readthrough efficacy

MotivationPremature termination codons (PTCs) are a major cause of genetic diseases, but the efficacy of therapeutic readthrough agents is highly context-dependent. While linear models have shown promise in predicting readthrough efficiency, they may not fully capture the complex, non-linear interactions between sequence context and drug activity. MethodsWe developed TransStop, a transformer-based pan-drug model, trained on a dataset of [~]5,400 PTCs and eight readthrough compounds. The model learns sequence representations and incorporates learnable embeddings to capture drug-specific effects, allowing a single model to predict efficacy for multiple drugs. ResultsOur model achieved a global R2=0.94 on a held-out test set. Visualizations of the learned embeddings revealed a deep understanding of biological principles, including the distinct clustering of stop codon types and grouping of drugs by mechanism of action. In silico saturation mutagenesis and epistasis analyses uncovered complex, non-additive sequence determinants of readthrough. We generated 32.7 million predictions across the human genome, covering all possible PTCs. Analysis of these genome-wide predictions revealed strong drug specializations for specific stop codon contexts and identified key areas of disagreement with previous models, particularly for UGA codons, where our model predicts a more effective drug in thousands of cases. The TransStop model represents a significant advancement in the prediction of translational readthrough efficiency. Its superior accuracy and the biological insights derived from its applications provide a powerful tool for guiding clinical trial design, drug development, and personalized patient treatment. Availability and ImplementationSource code: https://github.com/Dichopsis/TransStop. Model: https://huggingface.co/Dichopsis/TransStop.Genome-wide predictions: https://doi.org/10.5281/zenodo.16918476.

bioinformatics↗

bioRxiv2024

Critical assessment of missense variant effect predictors on disease-relevant variant data

Regular, systematic, and independent assessment of computational tools used to predict the pathogenicity of missense variants is necessary to evaluate their clinical and research utility and suggest directions for future improvement. Here, as part of the sixth edition of the Critical Assessment of Genome Interpretation (CAGI) challenge, we assess missense variant effect predictors (or variant impact predictors) on an evaluation dataset of rare missense variants from disease-relevant databases. Our assessment evaluates predictors submitted to the CAGI6 Annotate-All-Missense challenge, predictors commonly used by the clinical genetics community, and recently developed deep learning methods for variant effect prediction. To explore a variety of settings that are relevant for different clinical and research applications, we assess performance within different subsets of the evaluation data and within high-specificity and high-sensitivity regimes. We find strong performance of many predictors across multiple settings. Meta-predictors tend to outperform their constituent individual predictors; however, several individual predictors have performance similar to that of commonly used meta-predictors. The relative performance of predictors differs in high-specificity and high-sensitivity regimes, suggesting that different methods may be best suited to different use cases. We also characterize two potential sources of bias. Predictors that incorporate allele frequency as a predictive feature tend to have reduced performance when distinguishing pathogenic variants from very rare benign variants, and predictors supervised on pathogenicity labels from curated variant databases often learn label imbalances within genes. Overall, we find notable advances over the oldest and most cited missense variant effect predictors and continued improvements among the most recently developed tools, and the CAGI Annotate-All-Missense challenge (also termed the Missense Marathon) will continue to assess state-of-the-art methods as the field progresses. Together, our results help illuminate the current clinical and research utility of missense variant effect predictors and identify potential areas for future development.

genetics↗

bioRxiv2022

CeGAL: revisiting a widespread fungal-specific TF family using an in silico error-aware approach to identify missing zinc cluster domains

Transcription factors (TF) regulate gene activity in eukaryotic cells by binding specific regions of genomic DNA. In fungi, the most abundant TF class contains a fungal-specific GAL4-like Zn2C6 DNA binding domain (DBD), while the second class contains another fungal-specific domain, known as fungal_trans or Middle Homology Domain (MHD), whose function remains largely uncharacterized. Remarkably, almost a third of MHD-containing TF in public sequence databases apparently lack DNA binding activity, since they are not predicted to contain a DBD. Here, we reassess the domain organization of these MHD-only proteins using an in silico error-aware approach. Our large-scale analysis of ~17000 MHD-only TF sequences showed that the vast majority (>90%) result from gene annotation errors, thus contradicting previous findings that the MHD-only TF are widespread in fungi. We show that they are in fact exceptional cases, and that the Zn2C6-MHD domain pair represents the canonical domain signature defining a new TF family composed of two fungal-specific domains. We call this family CeGAL, after the most characterized members: Cep3, whose 3D structure has been determined and GAL4, an archetypal eukaryotic TF. This definition should improve the classification of the Zn2C6 TF and provide critical insights into fungal gene regulatory networks. IMPORTANCEIn fungi, extensive efforts focus on genome-wide characterization of potential Transcription Factors (TFs) and their targets genes to provide a better understanding of fungal processes and a rational for transcriptional manipulation. The second most abundant families of fungal-specific TFs, characterized by a Middle Homology Domain, are major regulators of primary and secondary metabolisms, multidrug resistance and virulence. Remarkably, one third of these TFs do not have a DNA Binding Domain (DBD-orphan) and thus are excluded from genome-wide studies. This particularity has been the subject of debate for many years. By computationally inspecting the close genomic environment of about 20,000 DBD-orphan TFs from a wide range of fungal species, we reveal that more than 90% contained sequences encoding a zinc-finger DBD. This analysis implies that the arrays of DBD containing TFs and their control DNA-sequences in target genes need to be reconsidered and expands the combinatorial regulation degree of the crucial fungal processes controlled by this TF family.

bioinformatics↗

bioRxiv2022

IMPatienT: an integrated web application to digitize, process and explore multimodal patient data.

Medical acts, such as imaging, lead to the production of several medical text report that describes the relevant findings. This induces multimodality in patient data by linking image data to free-text and consequently, multimodal data have become central to drive research and improve diagnosis. However, the exploitation of patient data is challenging as the ecosystem of analysis tools is fragmented depending on the type of data (images, text, genetics), the task (processing, exploration) and domains of interest (clinical phenotype, histology). To address the challenges, we present IMPatienT (Integrated digital Multimodal PATIENt daTa), a simple, flexible and open-source web application to digitize, process and explore multimodal patient data. IMPatienT has a modular architecture to: (i) create a standard vocabulary for a domain, (ii) digitize and process free-text data, (iii) annotate images and perform image segmentation, and (iv) generate a visualization dashboard and perform diagnosis suggestions. We showcased IMPatienT on a corpus of 40 simulated muscle biopsy reports of congenital myopathy patients. As IMPatienT relies on a user-designed vocabulary, it can be adapted to any domain of research and can be used as a patient registry for exploratory data analysis (EDA). A demo instance of the application is available at https://impatient.lbgi.fr/.

bioinformatics↗

bioRxiv2020

Characterization of accessory genes in coronavirus genomes

The Covid19 infection is caused by the SARS-CoV-2 virus, a novel member of the coronavirus (CoV) family. CoV genomes code for a ORF1a / ORF1ab polyprotein and four structural proteins widely studied as major drug targets. The genomes also contain a variable number of open reading frames (ORFs) coding for accessory proteins that are not essential for virus replication, but appear to have a role in pathogenesis. The accessory proteins have been less well characterized and are difficult to predict by classical bioinformatics methods. We propose a computational tool GOFIX to characterize potential ORFs in virus genomes. In particular, ORF coding potential is estimated by searching for enrichment in motifs of the X circular code, that is known to be over-represented in the reading frames of viral genes. We applied GOFIX to study the SARS-CoV-2 and related genomes including SARS-CoV and SARS-like viruses from bat, civet and pangolin hosts, focusing on the accessory proteins. Our analysis provides evidence supporting the presence of overlapping ORFs 7b, 9b and 9c in all the genomes and thus helps to resolve some differences in current genome annotations. In contrast, we predict that ORF3b is not functional in all genomes. Novel putative ORFs were also predicted, including a truncated form of the ORF10 previously identified in SARS-CoV-2 and a little known ORF overlapping the Spike protein in Civet-CoV and SARS-CoV. Our findings contribute to characterizing sequence properties of accessory genes of SARS coronaviruses, and especially the newly acquired genes making use of overlapping reading frames.

bioinformatics↗

bioRxiv2020

Potential role of the X circular code in the regulation of gene expression

The X circular code is a set of 20 trinucleotides (codons) that has been identified in the protein-coding genes of most organisms (bacteria, archaea, eukaryotes, plasmids, viruses). It has been shown previously that the X circular code has the important mathematical property of being an error-correcting code. Thus, motifs of the X circular code, i.e. a series of codons belonging to X, which are significantly enriched in the genes, allow identification and maintenance of the reading frame in genes. X motifs have also been identified in many transfer RNA (tRNA) genes and in important functional regions of the ribosomal RNA (rRNA), notably in the peptidyl transferase center and the decoding center. Here, we investigate the potential role of X motifs as functional elements in the regulation of gene expression. Surprisingly, the definition of a simple parameter identifies several relations between the X circular code and gene expression. First, we identify a correlation between the 20 codons of the X circular code and the optimal codons/dicodons that have been shown to influence translation efficiency. Using previously published experimental data, we then demonstrate that the presence of X motifs in genes can be used to predict the level of gene expression. Based on these observations, we propose the hypothesis that the X motifs represent a new genetic signal, contributing to the maintenance of the correct reading frame and the optimization and regulation of gene expression. Author SummaryThe standard genetic code is used by (quasi-) all organisms to translate information in genes into proteins. Recently, other codes have been identified in genomes that increase the versatility of gene decoding. Here, we focus on the circular codes, an important class of genome codes, that have the ability to detect and maintain the reading frame during translation. Motifs of the X circular code are enriched in protein-coding genes from most organisms from bacteria to eukaryotes, as well as in important molecules in the gene translation machinery, including transfer RNA (tRNA) and ribosomal RNA (rRNA). Based on these observations, it has been proposed that the X circular code represents an ancestor of the standard genetic code, that was used in primordial systems to simultaneously decode a smaller set of amino acids and synchronize the reading frame. Using previously published experimental data, we highlight several links between the presence of X motifs in genes and more efficient gene expression, supporting the hypothesis that the X circular code still contributes to the complex dynamics of gene regulation in extant genomes.

bioinformatics↗

bioRxiv2017

EvoKEN: evolutionary knowledge extraction in networks

We introduce a multi-factorial, multi-level approach to build and explore evolutionary scenarios of complex protein networks. EvoKEN combines a unique formalism for integrating multiple types of data associated with network molecular components and knowledge extraction techniques for detecting cohesive/anomalous evolutionary processes. We analyzed known human pathway maps and identified perturbations or specializations at the local topology level that reveal important evolutionary and functional aspects of these cellular systems.

evolutionary biology↗