New computational models and learning algorithms
We are evaluating and validating new computational models for representing a mutation, as well as new machine learning algorithms, in particular bioinspired (genetic programming and neural networks/deep learning) to build robust, efficient and effective predictive models of the various consequences of protein mutations.
Sequence and evolutionary information
In addition the the structural information used in our predictive platform, we are also including sequence and evolutionary information in order to better analyse the effects of mutations that alter post-translational modification and localisation, and those mutations within disordered regions which, while often lack detailed structural information, play very important roles in protein interactions. Ultimately we would also like to be able to predict the effects of mutations based upon protein sequence.
Conformational dynamics
Proteins are dynamic molecules, and we have shown previously that how mutations alter the equilibrium between different conformations is important for understanding their role in disease and drug resistance. To address this we are using a combination of structural ensembles and coarse grained molecular dynamics to present a picture of the mutational effects across its conformational states.
Complex mutations
To date, ours and others methods have focused on understanding the effects of single point missense mutations. However using a simulated thermodynamic cycle, we are developing methods able to predict the effects of more complex mutations including insertions, deletions, alternative splicing events, multiple point mutations and heterozygous mutations.
Synonymous mutations
Our hypothesis and work is based on the combined theory of codon optimality and cotranslational folding, that the choice of codon directly affects the translation efficiency of the ribosome. Consequently, different codons give the nascent polypeptide chain varying amounts of time to explore the fold-space and as such the choice of codon directly affects the final structure of protein.