How to use motif analysis to predict binding sites in eucaryotes: Benchmarks of algorithms and some well-known applications.

Maximilian Hauessler (MSNC, Gif/Yvette)

The analysis of isolated words in non-coding regions has been used since almost a decade now. Too many different algorithms await the interested user and many results are never validated. However, there are success stories, despite the common problems when working on gene regulation and even a comparison of some approaches has been published. This talk will introduce the idea of motif prediction and discovery and presents some effective applications from the literature.

Kernels for gene regulatory regions

Jean Philippe Vert (Ecole des Mines de Paris, Centre for Computational Biology)

We present an approach that aims at the detection of conserved regulatory elements that can explain functions of genes. The approach is formulated in a supervised machine learning setting, where the problem is to predict the function of a gene from its promoter sequence by focusing of conserved motifs.
For that purpose we describe a hierarchy of motif-based positive definite kernels for multiple alignments of biological sequences. Each kernel can be thought of as a way to extract features from multiple alignments. The kernels incorporate progressively more information, with the most complex kernel accounting for a multiple alignment of orthologous regions, the phylogenetic tree relating the species, and the prior knowledge that relevant sequence patterns occur in conserved motif blocks. These kernels can be used in the presence of a library of known transcription factor binding sites, or de novo by iterating over all k-mers
of a given length. In the latter mode, a discriminative classifier built from such a kernel not only recognizes a given class of promoter regions, but as a side effect simultaneously identifies a collection of relevant, discriminative sequence motifs. We demonstrate the utility of the motif-based multiple alignment kernels by using a collection of aligned promoter regions from five yeast species to recognize classes of cell-cycle regulated genes.

Reference:
J.-P. Vert, R. Thurman and W. S. Noble, "Kernels for gene regulatory regions",
in Advances in Neural Information Processing Systems 18 (NIPS 2005), Y. Weiss, B. Schölkopf and J. Platt (Eds.), p.1401-1408, MIT Press, Cambridge, MA, 2006.

Analysis and annotation of the transcriptional regulatory sequences of higher eukaryotes: the point of view from the wet lab

Jean Imbert (Centre de Recherche en Cancérologie de Marseille – INSERM U599)

Un résumé plus complet est disponible en pièce jointe (pdf)

For genes that have been successfully delineated within the human genome sequence, most regulatory sequences that control their transcription remain to be elucidated. Hence, comprehensive identification of the cis-acting regulatory elements is one of the major challenges of genome biology. Pennachio and Rubin noted in 2001 that “Regulatory sequences constitute a small fraction of the roughly 95% of the human genome that does not encode proteins, but they determine the level, location and chronology of gene expression. Despite the importance of these non-coding sequences in gene regulation, our ability to identify and predict functions for this category of DNA is extremely limited” (1).

During my lecture, I will provide some examples how combining adequate in vitro and in vivo functional assays and some easily accessible bioinformatic tools helped me in deciphering the architecture of the complex regulatory regions characteristic of the human genes (14,15).

References:
1. Pennacchio, L. A., and E. M. Rubin. 2001. Genomic strategies to identify mammalian regulatory sequences.
Nat. Rev. Genet. 2: 100-109.
14. Kim, H. P., J. Imbert, and W. J. Leonard. 2006. Both integrated and differential regulation of components of the IL-2/IL-2 receptor system. Cytokine Growth Factors Rev, 17:349-366, 2006.
15. Grange, T., J. Imbert, and D. Thieffry. 2005. Epigenomics: large scale analysis of chromatin modifications and transcription factors/genome interactions. Bioessays 27: 1203-1205.