Learning the language of the genome: AI- and data-driven decoding of genetic variants and disease mechanisms

Chat with Graph Search

Ask any question about EPFL courses, lectures, exercises, research, news, etc. or try the example questions below.

DISCLAIMER: The Graph Chatbot is not programmed to provide explicit or categorical answers to your questions. Rather, it transforms your questions into API requests that are distributed across the various IT services officially administered by EPFL. Its purpose is solely to collect and recommend relevant references to content that you can explore to help you answer your questions.

Humans and animals both use acoustic signals for vocal communication. The advent of self-supervised learning (SSL) has enabled neural networks to learn robust and general feature representations through the intrinsic acoustic structure of input signals, wi ...

Machine learning models have made significant progress in load forecasting, but their forecast accuracy is limited in cases where historical load data is scarce. Inspired by the outstanding performance of large language models (LLMs) in computer vision and ...

Foundation models are highly versatile neural-network architectures capable of processing different data types, such as text and images, and generalizing across various tasks like classification and generation. Inspired by this success, we propose Foundati ...

Animals, including humans, interact with the external environment primarily through motion. Replicating their motor control skills in artificial embodied agents is a major objective of artificial intelligence research. This thesis presents a collection of ...

This thesis investigates methods to construct more capable models more efficiently, focusing on two aspects: improved architectures and optimization. We examine principled architectural modifications that reduce computational costs or introduce features fo ...

The dynamics of neuron populations commonly evolve on low-dimensional manifolds. Thus, we need methods that learn the dynamical processes over neural manifolds to infer interpretable and consistent latent representations. We introduce a representation lear ...

While the accurate description of redox reactions remains a challenge for first-principles calculations, it has been shown that extended Hubbard functionals (DFT+U+V) can provide a reliable approach, mitigating self-interaction errors, in materials with st ...

Variants in the GBA1 gene, encoding the lysosomal enzyme glucosylceramidase beta 1 (GCase), are linked to Parkinson’s disease (PD) and Gaucher disease (GD). Heterozygous variants increase PD risk, while homozygous variants lead to GD, a lysosomal storage d ...

Advances in generative models increase the need for sample quality assessment. To do so, previous methods rely on a pre-trained feature extractor to embed the generated samples and real samples into a common space for comparison. However, different feature ...

Functional connectivity patterns in the human brain, like the friction ridges of a fingerprint, can uniquely identify individuals. Does this “brain fingerprint” remain distinct even during Alzheimer’s disease (AD)? Using fMRI data from healthy and patholog ...

In the past few years, Machine Learning (ML) techniques have ushered in a paradigm shift, allowing the harnessing of ever more abundant sources of data to automate complex tasks. The technical workhorse behind these important breakthroughs arguably lies in ...

Goal: In patients with coronary artery disease, the prediction of future cardiac events such as myocardial infarction (MI) remains a major challenge. In this work, we propose a novel anatomy-informed multimodal deep learning framework to predict future MI ...

The design of nonviral protein-based delivery systems has gained significant attention as an alternative to viral vectors and lipid nanoparticles (LNPs) for gene therapy. While viral vectors offer efficient gene delivery, they present challenges such as im ...

Auditory research aims in general to lead to understanding of physiological processes. By contrast, the state of the art in automatic speech processing (notably recognition) is dominated by large pre-trained models that are meant to be used as black-boxes. ...

We identified a rare heterozygous germline loss-of-function variant in the tumor necrosis factor receptor-associated factor 2 (TRAF2) in a young adult patient diagnosed with medulloblastoma. This variant is located within the TRAF-C domain of the E3 ubiqui ...

Learning the language of the genome: AI- and data-driven decoding of genetic variants and disease mechanisms

Chat with Graph Search

Transferability of Learnt Speech Representations for Decoding Non-Human Vocal Communication

TimeGPT in load forecasting: A large time series model perspective

Foundation neural-networks quantum states as a unified Ansatz for multiple hamiltonians

Musculoskeletal motor control with reinforcement learning

Enhanced Architectures and Optimization Methods for Efficient Language Modeling

MARBLE: interpretable representations of neural population dynamics using geometric deep learning

Teaching oxidation states to neural networks

Allosteric Modulation of GCase Enhances Lysosomal Activity and Reduces ER Stress in GCase-Related Disorders

Assessing Sample Quality via the Latent Space of Generative Models

Fingerprints of brain disease: connectome identifiability in Alzheimer’s disease

Topics in statistical physics of high-dimensional machine learning

Anatomy-Informed Multimodal Learning for Myocardial Infarction Prediction

Protein Engineering and Directed Evolution for Nanocarrier Innovation

Training a Filter-Based Model of the Cochlea in the Context of Pre-Trained Acoustic Models

Germline loss-of-function variant in the E3 ubiquitin ligase TRAF2 in a young adult patient with medulloblastoma: a case report