Biology subjects

Marchal, A.

Publications and source records attributed to Marchal, A..

3 recordsLinked to original sources

bioRxiv2026

pyreml: a general-purpose, GPU-accelerated mixed model solver in Python

Linear mixed models (LMM) are a popular modeling and inference framework, widely used in bioinformatics and other scientific fields. To obtain an unbiased estimation of the variance components, an LMM has to be fitted using restricted maximum likelihood (REML), a computationally intensive step. Most REML software is closed source, and currently there is no Python library offering complete LMM functionalities, such as the specification of spatial and genomic variance structures, multivariate analyses, random regression and factors, and heteroscedastic residuals. Here we propose pyreml, an LMM solver that fills this gap, comes with a user-centered interface and speeds up calculations by running on GPU. We show that pyreml fits a genomic best linear unbiased prediction (GBLUP) model more than one order of magnitude (>12\times) faster than the current state-of-the-art, closed source REML software. We also show that \pyreml can estimate a latent factor-analytic variance structure, a challenging feature that no current open source REML solution implements. Finally, we illustrate the use of pyreml to analyze longitudinal data, and to perform spatial analysis. Anchored by 500+ functional tests assessing the equivalence of its outputs to those from reference software (lme4, rrBLUP and BLUPF90), we demonstrate pyreml to be a performant, reliable and open-source solution available for academic, research and industrial LMM users.

bioinformatics↗

bioRxiv2020

CD28 costimulatory domain protects against tonic signaling-induced functional impairment in CAR-Tregs

The use of chimeric antigen receptor (CAR)-engineered regulatory T cells (Tregs) has emerged as a promising strategy to promote immune tolerance in transplantation. However, in conventional T cells (Tconvs), CAR expression is often associated with tonic signaling resulting from ligand-independent baseline activation. Tonic signaling may cause CAR-T cell dysfunction, especially when the CAR structure incorporates the CD28 costimulatory domain (CSD) rather than the 4-1BB CSD. Here, we explored the impact of tonic signaling on human CAR-Tregs according to the type of CSD. Compared to CD28-CAR-Tregs, 4-1BB-CAR-Tregs showed enhanced proliferation and greater activation of MAP kinase and mTOR pathways but exhibited decreased lineage stability and reduced abilities to produce IL-10 and be restimulated through the CAR. Although both CAR-Treg populations were suppressive in vivo, cell tracking with bioluminescence imaging found longer persistence for CD28-CAR-Tregs than for 4-1BB-CAR-Tregs. This study demonstrates that CD28-CAR best preserves Treg function and survival in the context of tonic signaling, in contrast with previous findings for Tconvs. SUMMARYLamarthee et al investigated the impact of chimeric antigen receptor (CAR) tonic signaling on CAR-engineered Tregs according to the incorporated costimulatory domain (either CD28 or 4-1BB). CD28 ameliorated Treg stability, long-term survival and function compared to 4-1BB.

immunology↗

bioRxiv2018

Deciphering hybrid larch reaction norms using random regression

The link between phenotypic plasticity and heterosis is a broad fundamental question, with stakes in breeding. We report a case-study evaluating temporal series of wood ring traits of hybrid larch (Larix decidua x L. kaempferi and reciprocal) in relation to soil water availability. Growth rings record the tree plastic responses to past environmental conditions, and we used random regressions to estimate the reaction norms of ring width and wood density with respect to water availability. We investigated the role of phenotypic plasticity on the construction of hybrid larch heterosis and on the expression of its quantitative genetic parameters. The data came from an intra-/interspecific diallel mating design between both parental species. Progenies were grown in two environmentally contrasted sites, in France. Ring width plasticity with respect to water availability was confirmed, as all three taxa produced narrower rings under the lowest water availability. Hybrid larch appeared to be the most plastic taxon as its superiority over its parental species increased with increasing water availability. Despite the low heritabilities of the investigated traits, we found that the quantitative genetic parameters varied along the water availability gradient. Finally, by means of a complementary simulation, we demonstrated that random regression can be applied to model the reaction norms of non-repeated records of phenotypic plasticity bound by a family structure. Random regression is a powerful tool for the modeling of reaction norms in various contexts, especially perennial species.

genetics↗