arXiv subjects

Laurent Jacob

Publications and source records attributed to Laurent Jacob.

14 recordsLinked to original sources

arXiv2025

Likelihood-free inference of phylogenetic tree posterior distributions

Phylogenetic inference, the task of reconstructing how related sequences evolved from common ancestors, is a central objective in evolutionary genomics. The current state-of-the-art methods exploit probabilistic models of sequence evolution along phylogenetic trees, by searching for the tree maximizing the likelihood of observed sequences, or by estimating the posterior of the tree given the sequences in a Bayesian framework. Both approaches typically require to compute likelihoods, which is only feasible under simplifying assumptions such as independence of the evolution at the different positions of the sequence, and even then remains a costly operation. Here we present the first likelihood-free inference method for posterior distributions over phylogenies. It exploits a novel expressive encoding for pairs of sequences, and a parameterized probability distribution factorized over a succession of subtree merges. The resulting network provides well-calibrated estimates of the posterior distribution leading to more accurate tree topologies than existing methods, even under models amenable to likelihood computation. We further show that its edge against likelihood-based methods dramatically increases under models of sequence evolution with intractable likelihoods.

q-bio.PE↗

arXiv2022

Neural Networks beyond explainability: Selective inference for sequence motifs

Over the past decade, neural networks have been successful at making predictions from biological sequences, especially in the context of regulatory genomics. As in other fields of deep learning, tools have been devised to extract features such as sequence motifs that can explain the predictions made by a trained network. Here we intend to go beyond explainable machine learning and introduce SEISM, a selective inference procedure to test the association between these extracted features and the predicted phenotype. In particular, we discuss how training a one-layer convolutional network is formally equivalent to selecting motifs maximizing some association score. We adapt existing sampling-based selective inference procedures by quantizing this selection over an infinite set to a large but finite grid. Finally, we show that sampling under a specific choice of parameters is sufficient to characterize the composite null hypothesis typically used for selective inference-a result that goes well beyond our particular framework. We illustrate the behavior of our method in terms of calibration, power and speed and discuss its power/speed trade-off with a simpler data-split strategy. SEISM paves the way to an easier analysis of neural networks used in regulatory genomics, and to more powerful methods for genome wide association studies (GWAS).

q-bio.GN↗

arXiv2020

Convolutional Kernel Networks for Graph-Structured Data

We introduce a family of multilayer graph kernels and establish new links between graph convolutional neural networks and kernel methods. Our approach generalizes convolutional kernel networks to graph-structured data, by representing graphs as a sequence of kernel feature maps, where each node carries information about local graph substructures. On the one hand, the kernel point of view offers an unsupervised, expressive, and easy-to-regularize data representation, which is useful when limited samples are available. On the other hand, our model can also be trained end-to-end on large-scale data, leading to new types of graph convolutional neural networks. We show that our method achieves competitive performance on several graph classification benchmarks, while offering simple model interpretation. Our code is freely available at https://github.com/claying/GCKN.

stat.ML↗

arXiv2019

Recurrent Kernel Networks

Substring kernels are classical tools for representing biological sequences or text. However, when large amounts of annotated data are available, models that allow end-to-end training such as neural networks are often preferred. Links between recurrent neural networks (RNNs) and substring kernels have recently been drawn, by formally showing that RNNs with specific activation functions were points in a reproducing kernel Hilbert space (RKHS). In this paper, we revisit this link by generalizing convolutional kernel networks---originally related to a relaxation of the mismatch kernel---to model gaps in sequences. It results in a new type of recurrent neural network which can be trained end-to-end with backpropagation, or without supervision by using kernel approximation techniques. We experimentally show that our approach is well suited to biological sequences, where it outperforms existing methods for protein classification tasks.

stat.ML↗

arXiv2019

Anatomy of the vertebral column lymphatic network in mice

Cranial lymphatic vessels (LVs) are involved in transport of fluids, macromolecules and CNS immune responses. Little information about spinal LVs is available, because these delicate structures are embedded within vertebral tissues and difficult to visualize using traditional histology. Here we reveal an extended vertebral column LV network using three-dimensional imaging of decalcified iDISCO-clarified spine segments. Spinal LVs are metameric circuits exiting along spinal nerve roots and connecting to lymph nodes and the thoracic duct. They navigate in the epidural space and the dura mater around the spinal cord, and associate with leukocytes, peripheral dorsal root and sympathetic ganglia. Spinal LVs are VEGF-C-dependent and remodel extensively after spinal cord injury. They constitute an extension to cranial circuits for meningeal fluids, but also a route for perineural fluids and a link with peripheral immune and nervous circuits. Vertebral column LVs may be potential targets to improve the maintenance and repair of 32 spinal tissues as well as gatekeepers of CNS immunity.

q-bio.TO↗

arXiv2018

A Search for Refraction in Kepler Photometry of Gas Giants

Refraction can lead to a brightening just before ingress and just after egress of a transit, as light passes through the exoplanet's atmosphere and is refracted into our line of sight. Refraction just outside of transit has been seen and modeled in our own solar system during transits of Venus. For short-period planets, the model of Sidis and Sari (2010 ApJ,720,904) implies refraction peaks typically under 100 parts per million and comparable in duration to ingress and egress. Kepler photometry currently provides the best opportunity for detecting refraction. We search for the signature of refraction just outside of transit in Kepler photometry of 45 gas giants and firmly rule out the Sidis and Sari model for four candidates.

astro-ph.EP↗

arXiv2012

Correcting gene expression data when neither the unwanted variation nor the factor of interest are observed

When dealing with large scale gene expression studies, observations are commonly contaminated by unwanted variation factors such as platforms or batches. Not taking this unwanted variation into account when analyzing the data can lead to spurious associations and to missing important signals. When the analysis is unsupervised, e.g., when the goal is to cluster the samples or to build a corrected version of the dataset - as opposed to the study of an observed factor of interest - taking unwanted variation into account can become a difficult task. The unwanted variation factors may be correlated with the unobserved factor of interest, so that correcting for the former can remove the latter if not done carefully. We show how negative control genes and replicate samples can be used to estimate unwanted variation in gene expression, and discuss how this information can be used to correct the expression data or build estimators for unsupervised problems. The proposed methods are then evaluated on three gene expression datasets. They generally manage to remove unwanted variation without losing the signal of interest and compare favorably to state of the art corrections.

stat.AP↗

arXiv2011

Group Lasso with Overlaps: the Latent Group Lasso approach

We study a norm for structured sparsity which leads to sparse linear predictors whose supports are unions of prede ned overlapping groups of variables. We call the obtained formulation latent group Lasso, since it is based on applying the usual group Lasso penalty on a set of latent variables. A detailed analysis of the norm and its properties is presented and we characterize conditions under which the set of groups associated with latent variables are correctly identi ed. We motivate and discuss the delicate choice of weights associated to each group, and illustrate this approach on simulated data and on the problem of breast cancer prognosis from gene expression data.

stat.ML↗

arXiv2010

Gains in Power from Structured Two-Sample Tests of Means on Graphs

We consider multivariate two-sample tests of means, where the location shift between the two populations is expected to be related to a known graph structure. An important application of such tests is the detection of differentially expressed genes between two patient populations, as shifts in expression levels are expected to be coherent with the structure of graphs reflecting gene properties such as biological process, molecular function, regulation, or metabolism. For a fixed graph of interest, we demonstrate that accounting for graph structure can yield more powerful tests under the assumption of smooth distribution shift on the graph. We also investigate the identification of non-homogeneous subgraphs of a given large graph, which poses both computational and multiple testing problems. The relevance and benefits of the proposed approach are illustrated on synthetic data and on breast cancer gene expression data analyzed in context of KEGG pathways.

q-bio.QM↗

arXiv2010

Increasing stability and interpretability of gene expression signatures

Motivation : Molecular signatures for diagnosis or prognosis estimated from large-scale gene expression data often lack robustness and stability, rendering their biological interpretation challenging. Increasing the signature's interpretability and stability across perturbations of a given dataset and, if possible, across datasets, is urgently needed to ease the discovery of important biological processes and, eventually, new drug targets. Results : We propose a new method to construct signatures with increased stability and easier interpretability. The method uses a gene network as side interpretation and enforces a large connectivity among the genes in the signature, leading to signatures typically made of genes clustered in a few subnetworks. It combines the recently proposed graph Lasso procedure with a stability selection procedure. We evaluate its relevance for the estimation of a prognostic signature in breast cancer, and highlight in particular the increase in interpretability and stability of the signature.

stat.ML↗

arXiv2008

Clustered Multi-Task Learning: A Convex Formulation

In multi-task learning several related tasks are considered simultaneously, with the hope that by an appropriate sharing of information across tasks, each task may benefit from the others. In the context of learning linear functions for supervised classification or regression, this can be achieved by including a priori information about the weight vectors associated with the tasks, and how they are expected to be related to each other. In this paper, we assume that tasks are clustered into groups, which are unknown beforehand, and that tasks within a group have similar weight vectors. We design a new spectral norm that encodes this a priori assumption, without the prior knowledge of the partition of tasks into groups, resulting in a new convex optimization formulation for multi-task learning. We show in simulations on synthetic examples and on the IEDB MHC-I binding dataset, that our approach outperforms well-known convex methods for multi-task learning, as well as related non convex methods dedicated to the same problem.

cs.LG↗

arXiv2008

Virtual screening of GPCRs: an in silico chemogenomics approach

The G-protein coupled receptor (GPCR) superfamily is currently the largest class of therapeutic targets. \textit{In silico} prediction of interactions between GPCRs and small molecules is therefore a crucial step in the drug discovery process, which remains a daunting task due to the difficulty to characterize the 3D structure of most GPCRs, and to the limited amount of known ligands for some members of the superfamily. Chemogenomics, which attempts to characterize interactions between all members of a target class and all small molecules simultaneously, has recently been proposed as an interesting alternative to traditional docking or ligand-based virtual screening strategies. We propose new methods for in silico chemogenomics and validate them on the virtual screening of GPCRs. The methods represent an extension of a recently proposed machine learning strategy, based on support vector machines (SVM), which provides a flexible framework to incorporate various information sources on the biological space of targets and on the chemical space of small molecules. We investigate the use of 2D and 3D descriptors for small molecules, and test a variety of descriptors for GPCRs. We show fo instance that incorporating information about the known hierarchical classification of the target family and about key residues in their inferred binding pockets significantly improves the prediction accuracy of our model. In particular we are able to predict ligands of orphan GPCRs with an estimated accuracy of 78.1%.

q-bio.QM↗

arXiv2007

Kernel methods for in silico chemogenomics

Predicting interactions between small molecules and proteins is a crucial ingredient of the drug discovery process. In particular, accurate predictive models are increasingly used to preselect potential lead compounds from large molecule databases, or to screen for side-effects. While classical in silico approaches focus on predicting interactions with a given specific target, new chemogenomics approaches adopt cross-target views. Building on recent developments in the use of kernel methods in bio- and chemoinformatics, we present a systematic framework to screen the chemical space of small molecules for interaction with the biological space of proteins. We show that this framework allows information sharing across the targets, resulting in a dramatic improvement of ligand prediction accuracy for three important classes of drug targets: enzymes, GPCR and ion channels.

q-bio.QM↗

arXiv2007

Epitope prediction improved by multitask support vector machines

Motivation: In silico methods for the prediction of antigenic peptides binding to MHC class I molecules play an increasingly important role in the identification of T-cell epitopes. Statistical and machine learning methods, in particular, are widely used to score candidate epitopes based on their similarity with known epitopes and non epitopes. The genes coding for the MHC molecules, however, are highly polymorphic, and statistical methods have difficulties to build models for alleles with few known epitopes. In this case, recent works have demonstrated the utility of leveraging information across alleles to improve the performance of the prediction. Results: We design a support vector machine algorithm that is able to learn epitope models for all alleles simultaneously, by sharing information across similar alleles. The sharing of information across alleles is controlled by a user-defined measure of similarity between alleles. We show that this similarity can be defined in terms of supertypes, or more directly by comparing key residues known to play a role in the peptide-MHC binding. We illustrate the potential of this approach on various benchmark experiments where it outperforms other state-of-the-art methods.

q-bio.QM↗