arXiv subjects

Magali Richard

Publications and source records attributed to Magali Richard.

5 recordsLinked to original sources

arXiv2026

On the Promises and Limits of Multi-omics Integration for Deconvolution: The HADACA3 Benchmark

Understanding the cellular composition of complex tissues, such as tumors, is a key challenge in biology and medicine. A common approach, known as deconvolution, aims to estimate the cellular composition from bulk molecular measurements. With the growing availability of multiple types of molecular data, it is often assumed that combining data sources should improve deconvolution performance. Here, we present HADACA3, a community-driven benchmark designed to evaluate this assumption. We conducted a four-day collaborative competition followed by a large-scale computational benchmark, testing more than 250,000 analysis pipelines across nine datasets with matched DNA methylation (DNAm) and RNA profiles, representing a wide range of biological and experimental conditions. Our framework jointly evaluates the impact of preprocessing, feature selection, modeling, and integration strategies. We find that DNAm alone achieves the highest median performance across datasets, making it the most stable and reliable single-modality approach. However, multi-omics integration strategies can regularly achieve higher top performance in specific datasets and pipeline configurations. Among the tested strategies, late integration based on error-weighted averaging provides a strong and reliable baseline, while non-linear early integration methods, such as optimal transport, show promising results on real biological datasets. Overall, our results show that multi-omics integration does not systematically improve average performance over DNAm alone, but can improve best-case performance in specific settings. This highlights a trade-off between robustness and peak performance, and emphasizes the importance of aligning integration strategies with the statistical properties of the data. All data, code, and evaluation tools are publicly available to support reproducible research and future method development.

q-bio.QM↗

arXiv2026

Inferring cellular heterogeneity with mixture models for DNA methylation rates

Cellular heterogeneity is a hallmark of biological tissues and plays a central role in disease progression, diagnosis, and prognosis. Yet, accurately characterizing this heterogeneity from bulk molecular profiles remains challenging because observed signals arise from mixtures of multiple cell populations. Cell deconvolution aim to recover the relative abundance of constituent cell types from such heterogeneous measurements, but most existing approaches implicitly rely on restrictive assumptions on residual errors, including independence, homoscedasticity, and normality. These assumptions are rarely satisfied in omics data, which are inherently bounded and overdispersed. In this work, we show that whole-genome cell-type specific DNA methylation profiles exhibit latent group structures that can substantially impair deconvolution accuracy when ignored. We therefore propose a mixture of non-negative Beta regression models estimated through an Expectation-Maximization algorithm for DNA methylation rates. Our framework naturally incorporates a feature selection mechanism through mixture component identification, making component selection a critical step of the inference procedure. We further propose a dedicated criterion for component selection and assess the performance of the approach through an extensive comparative study across several in vitro benchmark datasets. Our results demonstrate that deconvolution accuracy is highly sensitive to latent component structure and show that explicitly modeling this heterogeneity yields substantial improvements over standard whole-genome deconvolution strategies. Altogether, this work establishes mixture modeling of DNA methylation data as a powerful new direction for robust and accurate cell deconvolution.

stat.AP↗

arXiv2024

Group lasso based selection for high-dimensional mediation analysis

Mediation analysis aims to identify and estimate the effect of an exposure on an outcome that is mediated through one or more intermediate variables. In the presence of multiple intermediate variables, two pertinent methodological questions arise: estimating mediated effects when mediators are correlated, and performing high-dimensional mediation analyses when the number of mediators exceeds the sample size. This paper presents a two-step procedure for high-dimensional mediation analyses. The first step selects a reduced number of candidate mediators using an ad-hoc lasso penalty. The second step applies a procedure we previously developed to estimate the mediated effects, accounting for the correlation structure among the retained candidate mediators. We compare the performance of the proposed two-step procedure with state-of-the-art methods using simulated data. Additionally, we demonstrate its practical application by estimating the causal role of DNA methylation in the pathway between smoking and rheumatoid arthritis using real data.

q-bio.QM↗

arXiv2024

AI Competitions and Benchmarks, Practical issues: Proposals, grant money, sponsors, prizes, dissemination, publicity

This chapter provides a comprehensive overview of the pragmatic aspects involved in organizing AI competitions. We begin by discussing strategies to incentivize participation, touching upon effective communication techniques, aligning with trending topics in the field, structuring awards, potential recruitment opportunities, and more. We then shift to the essence of community engagement, and into organizational best practices and effective means of disseminating challenge outputs. Lastly, the chapter addresses the logistics, exposing on costs, required manpower, and resource allocation for effectively managing and executing a challenge. By examining these practical problems, readers will gain actionable insights to navigate the multifaceted landscape of AI competition organization, from inception to completion.

cs.LG↗

arXiv2021

Codabench: Flexible, Easy-to-Use and Reproducible Benchmarking Platform

Obtaining standardized crowdsourced benchmark of computational methods is a major issue in data science communities. Dedicated frameworks enabling fair benchmarking in a unified environment are yet to be developed. Here we introduce Codabench, an open-source, community-driven platform for benchmarking algorithms or software agents versus datasets or tasks. A public instance of Codabench (https://www.codabench.org/) is open to everyone, free of charge, and allows benchmark organizers to compare fairly submissions, under the same setting (software, hardware, data, algorithms), with custom protocols and data formats. Codabench has unique features facilitating the organization of benchmarks flexibly, easily and reproducibly, such as the possibility of re-using templates of benchmarks, and supplying compute resources on-demand. Codabench has been used internally and externally on various applications, receiving more than 130 users and 2500 submissions. As illustrative use cases, we introduce 4 diverse benchmarks covering Graph Machine Learning, Cancer Heterogeneity, Clinical Diagnosis and Reinforcement Learning.

cs.LG↗