Publications

Effects of gene dosage on cognitive ability: A function-based association study across brain and non-brain processes
Guillaume Huguet
Thomas Renne
Cécile Poulain
Alma Dubuc
Kuldeep Kumar
Sayeh Kazem
Worrawat Engchuan
Omar Shanta
Élise Douard
Catherine Proulx
Martineau Jean-Louis
Zohra Saci
Josephine Mollon
Laura M. Schultz
Emma E.M. Knowles
Simon R. Cox
David Porteous
Gail Davies
Paul Redmond
Sarah E. Harris … (see 10 more)
Gunter Schumann
Guillaume Dumas
Aurélie Labbe
Zdenka Pausova
Tomáš Paus
Stephen W. Scherer
Jonathan Sebat
Laura Almasy
David C. Glahn
Sébastien Jacquemont
Copy-number variants (CNVs) that increase the risk for neurodevelopmental disorders also affect cognitive ability. However, such CNVs remain… (see more) challenging to study due to their scarcity, limiting our understanding of gene-dosage-sensitive biological processes linked to cognitive ability. We performed a genome-wide association study (GWAS) in 258,292 individuals, which identified—for the first time—a duplication at 2q12.3 associated with higher cognitive performance. We developed a functional-burden analysis, which tested the association between cognition and CNVs disrupting 6,502 gene sets biologically defined across tissues, cell types, and ontologies. Among those, 864 gene sets were associated with cognition, and effect sizes of deletion and duplication were negatively correlated. The latter suggested that functions across all biological processes were sensitive to either deletions (e.g., subcortical regions, postsynaptic) or duplications (e.g., cerebral cortex, presynaptic). Associations between non-brain tissues and cognition were driven partly by constrained genes, which may shed light on medical comorbidities in neurodevelopmental disorders.
2024-11-30
Cell Genomics (published)
doi.org
Ensemble machine learning to accelerate industrial decarbonization: Prediction of Hansen solubility parameters for streamlined chemical solvent selection
Eslam G. Al-Sakkari
Ahmed Ragab
Mostafa Amer
Olumoye Ajao
Marzouk Benali
Daria C. Boffito
Hanane Dagdougui
Mouloud Amazouz
2024-11-30
Digital Chemical Engineering (published)
doi.org
Predicting the Mathematics Literacy of Resilient Students from High‐performing Economies: A Machine Learning Approach
Yimei Zhang
Maria Cutumisu
2024-11-30
Studies in Educational Evaluation (published)
doi.org
Towards AI-designed genomes using a variational autoencoder
Natasha K. Dudek
Doina Precup
Genomes encode elaborate networks of genes whose products must seamlessly interact to support living organisms. Humans’ capacity to unders… (see more)tand these biological systems is limited by their sheer size and complexity. In this article, we develop a proof of concept framework for training a machine learning (ML) algorithm to model bacterial genome composition. To achieve this, we create simplified representations of genomes in the form of binary vectors that indicate the encoded genes, henceforth referred to as genome vectors. A denoising variational autoencoder was trained to accept corrupted genome vectors, in which most genes had been masked, and reconstruct the original. The resulting model, DeepGenomeVector, effectively captures complex dependencies in genomic networks, as evaluated by both qualitative and quantitative metrics. An in-depth functional analysis of a generated genome vector shows that its encoded pathways are interconnected, near complete, and ecologically cohesive. On the test set, where the model’s ability to reconstruct uncorrupted genome vectors was evaluated, Area Under the Receiver Operating Curve (AUROC) and F1 scores of 0.98 and 0.83, respectively, support the model’s strong performance. This article showcases the power of ML approaches for synthetic biology and highlights the possibility that artifical intelligence agents may one day be able to design genomes that animate carbon-based cells.
2024-11-30
Proceedings of the Royal Society B: Biological Sciences (published)
doi.org
Uhura: A Benchmark for Evaluating Scientific Question Answering and Truthfulness in Low-Resource African Languages
Edward Bayes
Israel Abebe Azime
Jesujoba Oluwadara Alabi
Jonas Kgomo
Tyna Eloundou
Elizabeth Proehl
Kai Chen
Imaan Khadir
Naome Etori
Shamsuddeen Hassan Muhammad
C. Mpanza
Igneciah Pocia Thete
Dietrich Klakow
David Ifeoluwa Adelani
Evaluations of Large Language Models (LLMs) on knowledge-intensive tasks and factual accuracy often focus on high-resource languages primari… (see more)ly because datasets for low-resource languages (LRLs) are scarce. In this paper, we present Uhura -- a new benchmark that focuses on two tasks in six typologically-diverse African languages, created via human translation of existing English benchmarks. The first dataset, Uhura-ARC-Easy, is composed of multiple-choice science questions. The second, Uhura-TruthfulQA, is a safety benchmark testing the truthfulness of models on topics including health, law, finance, and politics. We highlight the challenges creating benchmarks with highly technical content for LRLs and outline mitigation strategies. Our evaluation reveals a significant performance gap between proprietary models such as GPT-4o and o1-preview, and Claude models, and open-source models like Meta's LLaMA and Google's Gemma. Additionally, all models perform better in English than in African languages. These results indicate that LMs struggle with answering scientific questions and are more prone to generating false claims in low-resource African languages. Our findings underscore the necessity for continuous improvement of multilingual LM capabilities in LRL settings to ensure safe and reliable use in real-world contexts. We open-source the Uhura Benchmark and Uhura Platform to foster further research and development in NLP for LRLs.
2024-11-30
ArXiv (preprint)
doi.org
arxiv.org
Visual Modality Prompt for Adapting Vision-Language Object Detectors
Heitor Rapela Medeiros
Atif Belal
Srikanth Muralidharan
Éric Granger
Marco Pedersoli
The zero-shot performance of object detectors degrades when tested on different modalities, such as infrared and depth. While recent work ha… (see more)s explored image translation techniques to adapt detectors to new modalities, these methods are limited to a single modality and apply only to traditional detectors. Recently, vision-language detectors, such as YOLO-World and Grounding DINO, have shown promising zero-shot capabilities, however, they have not yet been adapted for other visual modalities. Traditional fine-tuning approaches tend to compromise the zero-shot capabilities of the detectors. The visual prompt strategies commonly used for classification with vision-language models apply the same linear prompt translation to each image making them less effective. To address these limitations, we propose ModPrompt, a visual prompt strategy to adapt vision-language detectors to new modalities without degrading zero-shot performance. In particular, an encoder-decoder visual prompt strategy is proposed, further enhanced by the integration of inference-friendly task residuals, facilitating more robust adaptation. Empirically, we benchmark our method for modality adaptation on two vision-language detectors, YOLO-World and Grounding DINO, and on challenging infrared (LLVIP, FLIR) and depth (NYUv2) data, achieving performance comparable to full fine-tuning while preserving the model's zero-shot capability. Our code is available at: https://github.com/heitorrapela/ModPrompt
2024-11-30
ArXiv (preprint)
doi.org
arxiv.org
Machine learning-enhanced immunopeptidomics applied to T-cell epitope discovery for COVID-19 vaccines
Kevin A. Kovalchik
David J. Hamelin
Peter Kubiniok
Benoîte Bourdin
Fatima Mostefai
Raphaël Poujol
Bastien Paré
Shawn M. Simpson
John Sidney
Éric Bonneil
Mathieu Courcelles
Sunil Kumar Saini
Mohammad Shahbazy
Saketh Kapoor
Vigneshwar Rajesh
Maya Weitzen
Jean-Christophe Grenier
Bayrem Gharsallaoui
Loïze Maréchal
Zhaoguan Wu … (see 9 more)
Christopher Savoie
Alessandro Sette
Pierre Thibault
Isabelle Sirois
Martin A. Smith
Hélène Decaluwe
Julie G. Hussin
Mathieu Lavallée-Adam
Etienne Caron
Next-generation T-cell-directed vaccines for COVID-19 focus on establishing lasting T-cell immunity against current and emerging SARS-CoV-2 … (see more)variants. Precise identification of conserved T-cell epitopes is critical for designing effective vaccines. Here we introduce a comprehensive computational framework incorporating a machine learning algorithm—MHCvalidator—to enhance mass spectrometry-based immunopeptidomics sensitivity. MHCvalidator identifies unique T-cell epitopes presented by the B7 supertype, including an epitope from a + 1-frameshift in a truncated Spike antigen, supported by ribosome profiling. Analysis of 100,512 COVID-19 patient proteomes shows Spike antigen truncation in 0.85% of cases, revealing frameshifted viral antigens at the population level. Our EpiTrack pipeline tracks global mutations of MHCvalidator-identified CD8 + T-cell epitopes from the BNT162b4 vaccine. While most vaccine epitopes remain globally conserved, an immunodominant A*01-associated epitope mutates in Delta and Omicron variants. This work highlights SARS-CoV-2 antigenic features and emphasizes the importance of continuous adaptation in T-cell vaccine development.
2024-11-27
Nature Communications (published)
doi.org
SPARO: Selective Attention for Robust and Compositional Transformer Encodings for Vision
Ankit Vani
Bac Nguyen
Samuel Lavoie
Ranjay Krishna
Aaron Courville
Selective attention helps us focus on task-relevant aspects in the constant flood of our sensory input. This constraint in our perception al… (see more)lows us to robustly generalize under distractions and to new compositions of perceivable concepts. Transformers employ a similar notion of attention in their architecture, but representation learning models with transformer backbones like CLIP and DINO often fail to demonstrate robustness and compositionality. We highlight a missing architectural prior: unlike human perception, transformer encodings do not separately attend over individual concepts. In response, we propose SPARO, a read-out mechanism that partitions encodings into separately-attended slots, each produced by a single attention head. Using SPARO with CLIP imparts an inductive bias that the vision and text modalities are different views of a shared compositional world with the same corresponding concepts. Using SPARO, we demonstrate improvements on downstream recognition, robustness, retrieval, and compositionality benchmarks with CLIP (up to +14% for ImageNet, +4% for SugarCrepe), and on nearest neighbors and linear probe for ImageNet with DINO (+3% each). We also showcase a powerful ability to intervene and select individual SPARO concepts to further improve downstream task performance (up from +4% to +9% for SugarCrepe) and use this ability to study the robustness of SPARO's representation structure. Finally, we provide insights through ablation experiments and visualization of learned concepts.
2024-11-27
Lecture Notes in Computer Science (published)
doi.org
arxiv.org
Automatic Segmentation of the Spinal Cord Nerve Rootlets
Jan Valosek
Theo Mathieu
Raphaëlle Schlienger
Olivia S. Kowalczyk
Julien Cohen-Adad
Precise identification of spinal nerve rootlets is relevant to delineate spinal levels for the study of functional activity in the spinal co… (see more)rd. The goal of this study was to develop an automatic method for the semantic segmentation of spinal nerve rootlets from T2-weighted magnetic resonance imaging (MRI) scans. Images from two open-access MRI datasets were used to train a 3D multi-class convolutional neural network using an active learning approach to segment C2-C8 dorsal nerve rootlets. Each output class corresponds to a spinal level. The method was tested on 3T T2-weighted images from datasets unseen during training to assess inter-site, inter-session, and inter-resolution variability. The test Dice score was 0.67 +- 0.16 (mean +- standard deviation across testing images and rootlets levels), suggesting a good performance. The method also demonstrated low inter-vendor and inter-site variability (coefficient of variation <= 1.41 %), as well as low inter-session variability (coefficient of variation <= 1.30 %) indicating stable predictions across different MRI vendors, sites, and sessions. The proposed methodology is open-source and readily available in the Spinal Cord Toolbox (SCT) v6.2 and higher.
2024-11-25
International Society for Magnetic Resonance in Medicine Scientific Meeting and Exhibition (unknown)
doi.org
arxiv.org
Trellis tree-based analysis reveals stromal regulation of patient-derived organoid drug responses
María Ramos Zapatero
Alexander Tong
James W. Opzoomer
Rhianna O’Sullivan
Ferran Cardoso Rodriguez
Jahangir Sufi
Petra Vlckova
Callum Baird Nattress
Xiao Qin
Jeroen Claus
Daniel Hochhauser
Smita Krishnaswamy
Christopher J. Tape
Patient-derived organoids (PDOs) can model personalized therapy responses; however, current screening technologies cannot reveal drug respon… (see more)se mechanisms or how tumor microenvironment cells alter therapeutic performance. To address this, we developed a highly multiplexed mass cytometry platform to measure post-translational modification (PTM) signaling, DNA damage, cell-cycle activity, and apoptosis in >2,500 colorectal cancer (CRC) PDOs and cancer-associated fibroblasts (CAFs) in response to clinical therapies at single-cell resolution. To compare patient- and microenvironment-specific drug responses in thousands of single-cell datasets, we developed "Trellis"-a highly scalable, tree-based treatment effect analysis method. Trellis single-cell screening revealed that on-target cell-cycle blockage and DNA-damage drug effects are common, even in chemorefractory PDOs. However, drug-induced apoptosis is rarer, patient-specific, and aligns with cancer cell PTM signaling. We find that CAFs can regulate PDO plasticity-shifting proliferative colonic stem cells (proCSCs) to slow-cycling revival colonic stem cells (revCSCs) to protect cancer cells from chemotherapy.
2024-11-25
Cell (published)
doi.org
Words Matter: Leveraging Individual Text Embeddings for Code Generation in CLIP Test-Time Adaptation
Shambhavi Mishra
Julio Silva-Rodríguez
Ismail Ben Ayed
Marco Pedersoli
Jose Dolz
Vision-language foundation models, such as CLIP, have shown unprecedented zero-shot performance across a wide range of tasks. Nevertheless, … (see more)these models may be unreliable under distributional shifts, as their performance is significantly degraded. In this work, we explore how to efficiently leverage class text information to mitigate these distribution drifts encountered by large pre-trained vision-language models (VLMs) during test-time inference. In particular, we propose to generate pseudo-labels for the test-time samples by exploiting generic class text embeddings as fixed centroids of a label assignment problem, which is efficiently solved with Optimal Transport. Furthermore, the proposed adaptation method (CLIP-OT) integrates a multiple template knowledge distillation approach, which replicates multi-view contrastive learning strategies in unsupervised representation learning but without incurring additional computational complexity. Extensive experiments on multiple popular test-time adaptation benchmarks presenting diverse complexity empirically show the superiority of CLIP-OT, achieving performance gains of up to 7% over recent state-of-the-art methods, yet being computationally and memory efficient.
2024-11-25
ArXiv (preprint)
doi.org
arxiv.org
Bidirectional Generative Pre-training for Improving Time Series Representation Learning
Ziyang Song
Qincheng Lu
Mike He Zhu
David L Buckeridge
Yuemei Li
2024-11-24
Proceedings of the 9th Machine Learning for Healthcare Conference (published)
doi.org
proceedings.mlr.press