Research
Dr. Ian Simpson
Senior Post-Doctoral Research Fellow
Centres for Integrative Physiology and Neuroscience
Institute for Adaptive and Neural Computation
University of Edinburgh
Introduction
In order to understand the processes that become derailed during human neurological disease and to formulate approaches to their treatment we need to learn much more about how neural structures are formed. This means not just a physical understanding of the anatomical processes of development, but also a molecular one. The cell biological journeys from multi-potential progenitor cells to terminally differentiated and functionally mature structures of the CNS and PNS are achieved by the execution of hugely complex, dynamic and inter-connected transcriptional programmes. These programmes remain largely unknown and represent an immense, but essential objective for neuroscience research.
Experimental approaches
Traditionally, the role of transcription factors in neurogenesis and neural function has been based on phenotypic observations of transgenic gene knockouts and/or naturally occurring null or hypomorphic alleles. This reductionist approach has laid the foundation for our understanding of how developing neural systems are patterned, specified and refined. However, for many of the transcription factors involved in neural development these phenotypes can be so severe that what we observe are in fact secondary defects far downstream from the primary function of the gene. We  use conditional gene knockout strategies to investigate the role of genes in neural development in a much more controlled way.
Studying neural development using conditional gene knockouts (cKOs)
It is now relatively straightforward to create spatio-temporal knock-outs of genes in mammals using the Cre-loxP binary transgenic system in which transgenic animals are generated in which the gene of choice is ‘primed’ for deletion by flanking critical parts of the gene with small loxP DNA sequences. When mice carrying these ‘primed’ alleles are crossed with mice expressing Cre recombinase under either  spatial and/or temporal control we can ‘delete’ the gene only in the areas where the Cre recombinase is expressed. We use cKO mice to selectively remove transcription factors such as Pax6 during development to study specific aspects of its developmental function including fate specification, differentiation, patterning, cell migration, cell-cell signalling,cell-adhesion and axon navigation.
Capturing regulation, transcriptome and proteome analysis of mouse and fruitfly nervous systems
(tbc)
Assaying molecular function of transcription factors in neural development
(tbc)
Computational approaches
Regulatory genomics of neurogenesis and the control of neural function
(tbc)
Comparative evolutionary modelling of gene regulation, module reuse and circuitry
(tbc)
Biological network inference
We are using both static and dynamic Bayesian inference algorithms to generate the regulatory structural frameworks that underpin our modelling approaches. We are applying these methods to the inference of transcription factor regulatory circuits in fly PNS development and mammalian synapse plasticity in learning and memory. In addition we are creating networks to infer functional relationships between clusters of interacting proteins in neural domain PPI datasets. Our objective is to build functional networks that allow us to connect both transcriptional and translational data into pan-regulatory models.
Modeling synaptic complexity in learning and memory
By integrating novel and existing proteome and protein-protein interaction data we can assemble functional sub-networks that help us to understand the control systems regulating biological processes. One way in which we can integrate gene expression data with such models is by analysing sub-networks that are enriched in transcription factors. Using direct binding data and predictive sequence motifs for TF binding we can make predictions about likely targets for these factors and integrate them into our pan-regulatory networks.
Computational method development to study genome scale data in biology
MSc. Informatics projects 2011 (UoE restricted)