Mike Gilchrist group

Gene regulatory networks in early development

Embryo development is a complex and tightly controlled process, with a remarkably precise outcome. The underlying control system is only partly understood. Typically, transcription factors regulate the expression of individual genes, and the many relationships between transcription factors and their target genes combine to make gene regulatory networks. Our aim is to elucidate these networks using molecular and computational tools developed in the last few years that enable a systematic and large-scale approach.

We plan to use these methods to study the timing and localisation of gene expression in developmental model systems such as Xenopus. In particular we will be analysing time based profiles of gene expression generated by massively parallel, deep-sequencing technology, and combining this with expression co-localisation data derived from the computational comparison of in situ expression images, in order to generate large numbers of candidate relationships between genes. These relationships will then be validated and characterised, using both bioinformatic and experimental methods, and from these we can build more extensive and informative models of the gene regulatory networks controlling early development.

Embryonic gene expression profiles for Xenopus tropicalis derived from existing EST data

Embryonic gene expression profiles for Xenopus tropicalis derived from existing EST data

Click image to view at full-size

Using cDNA library data and gene-clustered ESTs, we can reconstruct the time-based behaviour of gene expression in frog embryos from the fertilised egg (stage 1) through to late metamorphosis. Here we see a clear distinction between maternal (upper profiles) and early zygotic (lower profiles) mRNAs, with shorter and longer persistence times. High-throughput sequencing technology will yield a hundred times better resolution and should enable clear inferences to be drawn about the progression of the embryonic transcriptional program.

Clustering in situ image data to extract candidate gene relationships.

Clustering in situ image data to extract candidate gene relationships.

Click image to view at full-size

Using computerised representation of embryonic gene expression patterns, we can cluster the images using a similarity metric, looking for pairs of genes with highly congruent expression patterns at the same stage of development. These co-localised genes are likely to have some direct or indirect functional relationship, and this can be further dissected using a combination of bioinformaticand experimental techniques.

Selected publications

Gilchrist group

Dr Mike Gilchrist

Mike Gilchrist
mgilchr@nimr.mrc.ac.uk

Recent publications

Our research themes

Click links to view others working on these themes

Top of page

© MRC National Institute for Medical Research
The Ridgeway, Mill Hill, London NW7 1AA