Himel Mallick
Principal Investigator (Tenure-track Faculty)
Research Interests
Make your next move.
Prepare your application
Review drafts before you send them.
Explore related searches
Articles (18)
Year:
2026
Year:
2026
Year:
2026
Year:
2026
Year:
2026
Microbial community analysis typically involves determining which microbial features are associated with properties such as environmental or health phenotypes. This task is impeded by data characteristics, including sparsity (technical or biological) and compositionality. Here we introduce MaAsLin 3 (microbiome multivariable associations with linear models) to simultaneously identify both abundance and prevalence relationships in microbiome studies with modern, potentially complex designs. MaAsLin 3 can newly account for compositionality either experimentally (for example, quantitative PCR or spike-ins) or computationally, and it expands the range of testable biological hypotheses and covariate types. On a variety of synthetic and real datasets, MaAsLin 3 outperformed state-of-the-art differential abundance methods, and when applied to the Inflammatory Bowel Disease Multi-omics Database, MaAsLin 3 corroborated previously reported associations, identifying 77% with feature prevalence rather than abundance. In summary, MaAsLin 3 enables researchers to identify microbiome associations more accurately and specifically, especially in complex datasets.
Year:
2026
S ummary There is an overwhelmingly large body of literature and numerous algorithms already available on “multimodal AI”, based on different modeling and fusion paradigms. As multimodal AI becomes fully integrated into healthcare applications, the ability to say “I’m not sure” or “I don’t know” when uncertain is a necessary capability for safe clinical deployment, requiring the augmentation of current multimodal AI techniques. Our goal in this work is not to add yet another method to the growing toolbox of multimodal AI, but rather to (i) clarify how existing supervised multimodal AI methods can be understood through the lens of uncertainty quantification, and (ii) develop a unified framework that provides valid and interpretable uncertainty estimates for multimodal AI practitioners. We introduce Coracle, a conformalized framework for multimodal AI that seamlessly adapts to early, late, and intermediate fusion. Coracle provides theoretical marginal confidence guarantees and achieves valid finite-sample coverage without relying on distributional assumptions. Through extensive simulations and analyses of diverse biomedical multiview datasets, we demonstrate that Coracle consistently outperforms existing non-conformalized methods in both uncertainty estimation and calibration. By equipping published multiview models with valid, interpretable confidence measures, Coracle advances the development of trustworthy, uncertainty-aware statistical tools for biomedical decision support. Coracle is publicly available as an open-source R package at https://github.com/himelmallick/Coracle .
Year:
2026
We reexamine GOOOGLE, the group‐regularized zero‐inflated negative binomial (ZINB) approach of Chatterjee et al. We show that in the released implementation, the tuning parameter is selected using a Bayesian information criterion (BIC) computed on a Gaussian surrogate. Because the unpenalized model fits this surrogate exactly (with zero residual sum of squares), the BIC always favors essentially unpenalized solutions and fails to induce group sparsity. This results in zero group specificity in simulations mirroring the original paper's design. We demonstrate that this issue is resolved by selecting the tuning parameter using the true ZINB log‐likelihood. Furthermore, we propose the fully iterative group broken adaptive ridge (grBAR) estimator as a more robust alternative. Our open‐source R package, group regularization algorithms for zero‐inflated models (GRAZIMs), provides these tools to enable reliable group selection in ZINB models.
Year:
2025
The expansion of microbiome research has led to the accumulation of interlinked datasets encompassing versatile taxonomic and functional assays. The analysis of increasingly large and heterogeneous multi-modal microbiome data would benefit from unified approaches supporting the design of modular data science workflows through interoperable methods. The Bioconductor project has recently developed an optimized statistical programming framework for multi-assay data integration. Building on this foundation, we introduce a community-developed open source ecosystem for microbiome data science. In contrast to the previous alternatives, the methodology is specifically designed to support joint analysis of hierarchical, interlinked, and heterogeneous multi-table datasets that are increasingly common in modern microbiome research. This data science ecosystem encompasses open data, methods, tutorials, and an active online community. These resources support standardized and reproducible data wrangling, joint analysis, and reporting. We have detailed the functionality and usage in the online book https://microbiome.github.io/OMA , which offers guidance for prospective users and contributors.
Year:
2025
Bayesian multimodal models that integrate data from multiple sources, studies, or modalities have garnered considerable attention in recent years. However, existing methods either rely on computationally expensive Markov chain Monte Carlo (MCMC) schemes or adopt variational approaches that forgo principled uncertainty quantification. To address this limitation and cater to practical needs, we abandon the MCMC framework and turn to resampling-based posterior inference for multimodal integration. Our method, Bayesian Cooperative Learning (BayesCOOP), embeds fast maximum a posteriori (MAP) estimation within a Bayesian bootstrap, combining a novel jittered group spike-and-slab prior with an efficient expectation–maximization (E M) coordinate descent algorithm under randomly weighted data perturbations. Averaging posterior summaries (MAP estimates) across bootstrap replicates yields approximate posterior samples that retain Bayesian interpretability while avoiding the computational burden of traditional sampling-based inference. We establish theoretical connections between BayesCOOP’s pseudo-posterior averaging and posterior contraction principles, demonstrating near-optimal posterior consistency under sparsity. Extensive simulation studies and analyses of pregnancy multi-omics datasets demonstrate that BayesCOOP substantially outperforms state-of-the-art early, intermediate, and late fusion approaches in estimation, prediction, and uncertainty assessment. The open-source implementation of BayesCOOP is available at https://github.com/himelmallick/BayesCOOP .
Year:
2025
Year:
2025
Year:
2025
Year:
2025
Year:
2025
Year:
2024
Year:
2024
Year:
2024
Year:
2024
Collaborators (17)
Levi Waldron
Professor
New York University
Ali Rahnavard
Associate Professor
George Washington University
Jacob Nearing
Harvard University
Arindam Fadikar
Computational Statistician
Argonne National Laboratory
Khemlal Nirmalkar
Research Scientist
Arizona State University
Curtis Huttenhower
Harvard University
Suvo Chatterjee
Assistant Professor (Tenure-track Faculty)
Indiana State University
Christian Lorenz Müller
Professor
Ludwig-Maximilians-Universität München
Aura Raulo
University of Oxford
Aki Havulinna
Senior Research Fellow
University of Helsinki
Katariina Pärnänen
Group leader, Academy Research Fellow
University of Helsinki
Emmanuel Ogundimu
Lecturer
University of Oxford
Leo Lahti
Associate Professor
University of Turku
Kate Solpari
Weill Cornell Medicine
James Adams
President's Professor
Arizona State University
Robert Brown
Gladys and Roland Harriman Professor of Medicine
Joan and Sanford I Weill Medical College of Cornell University
Rosa Krajmalnik-Brown
Arizona State University
Make your next move.
Prepare your application
Review drafts before you send them.