Integration of enhanced protein function prediction with experimental studies of fertilisation in Plasmodium - a wet/dry study

This proposal is to conduct a wet/dry study that will benefit from the synergies of a computational group developing novel bioinformatics strategies for protein function prediction working closely with an experimental group requiring these methodologies to prioritise wet studies to extend our understanding of fertilisation in Plasmodium. Genome projects are determining the sequences of proteins in numerous species including, man, model mammals, plants (including those of relevance to agriculture), and pathogens of human, animals and plants. Central to the understanding and exploitation of this information is the assignment of function to the proteins. A range of computational approaches have been developed to assign function to proteins. Several approaches, including one developed in our group, use information from the sequence and the 3D protein structure. Other approaches use information at a systems level / which genes are turned on or off together (transcriptomics, proteomic) and which proteins interact (interactomic). This proposal is to develop an enhanced method of function prediction that integrates information from sequences, structures, transcriptomes , proteomes and interactomes. The computer program will be made available to the academic community via a web server. We will also take part in international blind trials of function prediction. This general development of software will be targeted at understanding the fertilisation of Plasmodium gametes. One particular species of Plasmodium, (Plasmodium falciparum) is the parasite that is responsible for the majority of malarial deaths. The disease presents a risk to 40% of the world's population and is responsible for c. 400 million cases and 3 million deaths annually world wide. Plasmodium is transmitted from person to person by the bite of an anopheline mosquito. In addition, Plasmodium, and notably Plasmodium berghei (a parasite of rodents, that is not pathogenic to man) has become a model organism for study of parasite/host interactions because of the importance of understanding the molecular basis of malaria. The proteins in the gamete of P. berghei have been recently characterised in our laboratory. To direct our proposed experimental studies, the enhanced bioinformatics tools for function prediction will be applied to these gamete proteins. In addition, the bioinformatics tools will be used to improve the functional annotation of Plasmodium sequences. Our results will be added to the community databases (Gene DB and PlamoDB) describing the annotation of Plasmodium.

1. Evaluation and application of available function prediction tools to Plasmodium. The bioinformatician and the biologist will jointly annotate the proteome of the male gamete of Plasmodium, i.e. a wet/dry approach. We will also apply our software to the entire Plasmodium proteome and add our annotation to community databases (GeneDB and PlasmoDB). 2. Development and dissemination of function prediction from domain combination. We will identify each component domain combination (using sequence and enhanced fold recognition) and use machine learnt (support vector) rules to provide a set of possible functions for the unknown protein. We will take part in international blind testing of function prediction. 3. Development and dissemination of an integrated approach for function prediction. We will integrate the results from a range of other prediction approaches including enhanced sequence-based function prediction, co-expression data analysis and interactome data.. 4. Experimental characterisation in P. berghei of male gamete proteins involved in fertilisation directed by the bioinformatics analysis using: 4.1 Gene knockout - Two independent clones of transgenic parasites arising from 2 rounds of drug selection will be subjected to our routine analysis for their ability to make gametocytes, to exflagellate (make male gametes), to produce ookinetes in vitro, and to produce oocysts in vivo. 4.2 Protein localization - This will be approached using i) a robust method for gfp- and/or c-myc tagging malarial proteins and ii) confirm these studies by locating the native protein with antibodies raised to protein domains expressed in E.coli . 4.3 Protein complex detection - Hypotheses will be tested by using the antibodies and tagged parasite combinations described in step 5.2, to attempt to identify by co-precipitation proteins complexed to the target molecule. We will determine whether the complex is formed within the male gamete or between gamete sexes. Co-funded by EPSRC.