Freemont Lab

Conserved MITE repeats in gene clusters

Sequence repeats have been found in the promoter regions of the Sad gene cluster in oat by the lab of Anne Osbourn and were indentified as belonging to a MITE family. It is hypothesised that these gene clusters may be implicated in establishing the Sad gene cluster and/or in its regulation. In collaboration with the Osbourn lab and Derek Huntley at Imperial College, we are using bioinformatics tools to identify similar repeats in the Arabidopsis genome.

Sequence clustering algorithm development

A novel probabilistic sequence clustering algorithm has been developed which identifies statistically significant populations in sequence alignments and can then be used to generate further sequences. No prior assumptions are made about the sequence composition and requires no extra parameters such as amino acid transition matrices. For these reasons it can be used with mixtures of DNA and protein sequences, potentially providing insight into interactions between the two. This method is being applied to recombinase proteins and associated recombination core site sequences.

Computational protein loop design methods

A novel ab initio method to explore backbone loop conformations has been developed which has comparable performance to methods that explicitly insert fragments of known protein structure in standard loop modelling benchmarks. This method uses a coarse-grained potential energy function to sample conformational space with the advantage that (functional geometric) constraints can be incorporated into the search which is not possible to do with fragment insertion methods. This method potentially enables multiscale computational protein design methods with large-scale backbone sampling. This work will have uses in computationally redesigning the sequence specificity of recombinase proteins and is being experimentally validated in collaboration with the lab of James Murray at Imperial College.

Recombinase sequence specificity redesign

Work is being carried out to investigate re-engineering recombinase proteins to change integration site specificity using computational protein design methods.

Computational tools for DNA assembly

DNA assembly methods are a key enabling technology for synthetic biology but are currently a major bottleneck in the prototyping and production of designs. Multiple methods allow DNA parts to be assembled into new genes, complex networks and novel pathways, and recent methods have been used to construct entire synthetic genomes.

In collaboration with the labs of Tom Ellis and Geoff Baldwin at Imperial College and their PhD student Arturo Casini, we are developing computational algorithms to enable large scale DNA assembly.

GUI web-based parts drag-and-drop compositional design tool

An initial prototype GUI drag-and-drop tool has been developed to allow users to assemble synthetic biology designs in a web site.


  • Prof. Paul Freemont
  • Dr. James MacDonald

Related publications

  • Computational design approaches and tools for synthetic biology. James T. MacDonald, Chris Barnes, Richard I. Kitney, Paul S. Freemont and Guy-Bart V. Stan. Integr. Biol., 2011, 3:97-108.  [download pdf] [pubmed] [journal]