Julien Chiquet

Early allopolyploid evolution in the post-neolithic Brassica napus oilseed genome

The genomic origins of rape oilseed Many domesticated plants arose through the meeting of multiple genomes through hybridization and genome doubling, known as polyploidy. Chalhoub et al. sequenced the polyploid genome of Brassica napus, which originated from a recent combination of two distinct genomes approximately 7500 years ago and gave rise to the crops of rape oilseed (canola), kale, and rutabaga. B. napus has undergone multiple events affecting differently sized genetic regions where a gene from one progenitor species has been converted to the copy from a second progenitor species. Some of these gene conversion events appear to have been selected by humans as part of the process of domestication and crop improvement. Science, this issue p. 950 The polyploid genome of oilseed rape exhibits evolution through homologous gene conversion. Oilseed rape (Brassica napus L.) was formed ~7500 years ago by hybridization between B. rapa and B. oleracea, followed by chromosome doubling, a process known as allopolyploidy. Together with more ancient polyploidizations, this conferred an aggregate 72× genome multiplication since the origin of angiosperms and high gene content. We examined the B. napus genome and the consequences of its recent duplication. The constituent An and Cn subgenomes are engaged in subtle structural, functional, and epigenetic cross-talk, with abundant homeologous exchanges. Incipient gene loss and expression divergence have begun. Selection in B. napus oilseed types has accelerated the loss of glucosinolate genes, while preserving expansion of oil biosynthesis genes. These processes provide insights into allopolyploid evolution and its relationship with crop domestication and improvement.

Inferring multiple graphical structures

Gaussian Graphical Models provide a convenient framework for representing dependencies between variables. Recently, this tool has received a high interest for the discovery of biological networks. The literature focuses on the case where a single network is inferred from a set of measurements. But, as wetlab data is typically scarce, several assays, where the experimental conditions affect interactions, are usually merged to infer a single network. In this paper, we propose two approaches for estimating multiple related graphs, by rendering the closeness assumption into an empirical prior or group penalties. We provide quantitative results demonstrating the benefits of the proposed approaches. The methods presented in this paper are embeded in the R package simone from version 1.0-0 and later.

Inferring sparse Gaussian graphical models with latent structure

Our concern is selecting the concentration matrixs nonzero coefficients for a sparse Gaussian graphical model in a high-dimensional setting. This corresponds to estimating the graph of conditional dependencies between the variables. We describe a novel framework taking into account a latent structure on the concentration matrix. This latent structure is used to drive a penalty matrix and thus to recover a graphical model with a constrained topology. Our method uses an

Prevalence of gene expression additivity in genetically stable wheat allohexaploids

\ell_1

SIMoNe: Statistical Inference for MOdular NEtworks

penalized likelihood criterion. Inference of the graph of conditional dependencies between the variates and of the hidden variables is performed simultaneously in an iterative EM-like algorithm named SIMoNe (Statistical Inference for Modular Networks). Performances are illustrated on synthetic as well as real data, the latter concerning breast cancer. For gene regulation networks, our method can provide a useful insight both on the mutual influence existing between genes, and on the modules existing in the network.

Sparsity with sign-coherent groups of variables via the cooperative-Lasso

The reprogramming of gene expression appears as the major trend in synthetic and natural allopolyploids where expression of an important proportion of genes was shown to deviate from that of the parents or the average of the parents. In this study, we analyzed gene expression changes in previously reported, highly stable synthetic wheat allohexaploids that combine the D genome of Aegilops tauschii and the AB genome extracted from the natural hexaploid wheat Triticum aestivum. A comprehensive genome-wide analysis of transcriptional changes using the Affymetrix GeneChip Wheat Genome Array was conducted. Prevalence of gene expression additivity was observed where expression does not deviate from the average of the parents for 99.3% of 34,820 expressed transcripts. Moreover, nearly similar expression was observed (for 99.5% of genes) when comparing these synthetic and natural wheat allohexaploids. Such near-complete additivity has never been reported for other allopolyploids and, more interestingly, for other synthetic wheat allohexaploids that differ from the ones studied here by having the natural tetraploid Triticum turgidum as the AB genome progenitor. Our study gave insights into the dynamics of additive gene expression in the highly stable wheat allohexaploids.

Structured regularization for conditional Gaussian graphical models

SUMMARY The R package SIMoNe (Statistical Inference for MOdular NEtworks) enables inference of gene-regulatory networks based on partial correlation coefficients from microarray experiments. Modelling gene expression data with a Gaussian graphical model (hereafter GGM), the algorithm estimates non-zero entries of the concentration matrix, in a sparse and possibly high-dimensional setting. Its originality lies in the fact that it searches for a latent modular structure to drive the inference procedure through adaptive penalization of the concentration matrix. AVAILABILITY Under the GNU General Public Licence at http://cran.r-project.org/web/packages/simone/

Combining a relaxed EM algorithm with Occam's razor for Bayesian variable selection in high-dimensional regression

We consider the problems of estimation and selection of parameters endowed with a known group structure, when the groups are assumed to be sign-coherent, that is, gathering either nonnegative, nonpositive or null parameters. To tackle this problem, we propose the cooperative-Lasso penalty. We derive the optimality conditions defining the cooperative-Lasso estimate for generalized linear models, and propose an efficient active set algorithm suited to high-dimensional problems. We study the asymptotic consistency of the estimator in the linear regression setup and derive its irrepresentable conditions, which are milder than the ones of the group-Lasso regarding the matching of groups with the sparsity pattern of the true parameters. We also address the problem of model selection in linear regression by deriving an approximation of the degrees of freedom of the cooperative-Lasso estimator. Simulations comparing the proposed estimator to the group and sparse group-Lasso comply with our theoretical results, showing consistent improvements in support recovery for sign-coherent groups. We finally propose two examples illustrating the wide applicability of the cooperative-Lasso: first to the processing of ordinal variables, where the penalty acts as a monotonicity prior; second to the processing of genomic data, where the set of differentially expressed probes is enriched by incorporating all the probes of the microarray that are related to the corresponding genes.

A variable selection approach in the multivariate linear model: an application to LC-MS metabolomics data

Conditional Gaussian graphical models are a reparametrization of the multivariate linear regression model which explicitly exhibits (i) the partial covariances between the predictors and the responses, and (ii) the partial covariances between the responses themselves. Such models are particularly suitable for interpretability since partial covariances describe direct relationships between variables. In this framework, we propose a regularization scheme to enhance the learning strategy of the model by driving the selection of the relevant input features by prior structural information. It comes with an efficient alternating optimization procedure which is guaranteed to converge to the global minimum. On top of showing competitive performance on artificial and real datasets, our method demonstrates capabilities for fine interpretation, as illustrated on three high-dimensional datasets from spectroscopy, genetics, and genomics.

Fast Tree Inference With Weighted Fusion Penalties

We address the problem of Bayesian variable selection for high-dimensional linear regression. We consider a generative model that uses a spike-and-slab-like prior distribution obtained by multiplying a deterministic binary vector, which traduces the sparsity of the problem, with a random Gaussian parameter vector. The originality of the work is to consider inference through relaxing the model and using a type-II log-likelihood maximization based on an EM algorithm. Model selection is performed afterwards relying on Occams razor and on a path of models found by the EM algorithm. Numerical comparisons between our method, called spinyReg, and state-of-the-art high-dimensional variable selection algorithms (such as lasso, adaptive lasso, stability selection or spike-and-slab procedures) are reported. Competitive variable selection results and predictive performances are achieved on both simulated and real benchmark data sets. An original regression data set involving the prediction of the number of visitors of the Orsay museum in Paris using bike-sharing system data is also introduced, illustrating the efficiency of the proposed approach. The R package spinyReg implementing the method proposed in this paper is available on CRAN.