Statistical design of interactions between proteins that are both novel and specific
Lead Research Organisation:
University of Cambridge
Department Name: Chemistry
Abstract
The ability of proteins to associate with each other in specific interactions is crucial to the functioning of vast multitudes of biological processes. In the crowded molecular soup that makes up the intracellular environment, a single protein will encounter many other proteins with which they could potentially interact. The interactions that help cells reproduce and survive may range from fleeting connections to the formation of long lasting complexes; however it is the specificity of these interactions that allows the exquisite levels of regulation observed in so many biological systems. I will focus on how proteins evolve specific interactions, and how we can elucidate interaction partners from the vast amounts of genomic data currently being generated. Each and every cell is packed with proteins, the interactions between these proteins form the foundation of almost all cellular processes. I am interested in the ways that proteins fit together to form complexes, and the constraints that complex formation imposes on the evolution of the sequences of interaction partners. If one amino acid on the surface of protein A changes shape, in order for protein B to bind to protein A an amino acid on the surface of B may also have to change - this is an example of a compensatory mutation. I will develop statistical and mathematical analyses that extract correlations between different residues of a protein of interest from alignments of orthologous and paralogous proteins. Detecting compensatory mutations in sequence data provides information about protein structure and function, and the specificity of protein-protein interactions. The specificity of binding interactions is crucial to the proper functioning of cellular processes, and determining the evolutionary rules that govern the specificity of protein interactions will inform the rational design of novel specific interactions between proteins. Our ability to understand and thus engineer the molecular determinants of specificity is vital to efforts to design and engineer effective drugs and other bio-molecules. There are numerous potential applications across a range of different industries, for example synthetic biology approaches to energy production, and the development of biomimetic materials.
Planned Impact
I will make every effort to disseminate both the nature and results of my work beyond the confines of the international academic community. It is likely that there would be a high level of interest from industry in the ability to use statistical analyses to design novel interaction specificities. The Technology Transfer Group in the MRC Head Office is directly responsible for the management of exploitation of results from the council's laboratories and works in partnership with scientists to identify opportunities and develop and execute exploitation strategies. I would work closely with this group to ensure the protection of intellectual property. This might involve the patent process, to secure a proprietary position and thereby enhance the value of the opportunity for a prospective industrial partner. However, this process can be long, expensive, and often arduous and so will not be embarked upon lightly. I would also be interested in embarking on industrially funded collaborations. I am also happy to ensure that novel materials produced by this research are distributed to fellow academics under an appropriate Materials Transfer Agreement ensuring that the recipient does not commercially exploit without reference to the originator. The MRC-LMB has extensive experience of managing and exploiting results from the laboratory, and I would seek to benefit from this expertise. For example, antibody engineering technology was pioneered at the MRC-LMB, and the method of humanization of monoclonal antibodies through engineering the proteins was subsequently protected by a patent. Start up companies have also been founded by academics at the MRC-LMB in order to fully realize the potential of their scientific results. For example Cambridge Antibody Technology Ltd. (CAT) has disseminated a phage antibody screening technology that allows in vitro selection of novel antibodies from human repertoires to the wider public.
People |
ORCID iD |
Lucy Colwell (Principal Investigator) |
Publications
Townsend PD
(2015)
The Role of Protein-Ligand Contacts in Allosteric Regulation of the Escherichia coli Catabolite Activator Protein.
in The Journal of biological chemistry
Tesileanu T
(2014)
Protein sectors: statistical coupling analysis versus conservation
Tesileanu T
(2015)
Protein sectors: statistical coupling analysis versus conservation.
in PLoS computational biology
Perica T
(2012)
The emergence of protein complexes: quaternary structure, dynamics and allostery. Colworth Medal Lecture.
in Biochemical Society transactions
Liberles DA
(2012)
The interface of protein structure, protein biophysics, and molecular evolution.
in Protein science : a publication of the Protein Society
Hopf TA
(2012)
Three-dimensional structures of membrane proteins from genomic sequencing.
in Cell
Dwyer RS
(2013)
Predicting functionally informative mutations in Escherichia coli BamA using evolutionary covariance analysis.
in Genetics
Colwell LJ
(2014)
Conservation weighting functions enable covariance analyses to detect functionally important amino acids.
in PloS one
Colwell L
(2014)
Feynman-Hellmann Theorem and Signal Identification from Sample Covariance Matrices
in Physical Review X
Beh LY
(2012)
A core subunit of Polycomb repressive complex 1 is broadly conserved in function but not primary sequence.
in Proceedings of the National Academy of Sciences of the United States of America
Description | Working as part of a team of international collaborators I found that we are able to identify statistical patterns in the mutations of amino acids within protein sequences, and use this information to predict the three dimensional structure of the folded molecule. We use a Bayesian inference approach to analyze large sequence alignments. We also find that amino acid residues whose mutation patterns are correlated provide information about alternative conformations of individual proteins, in addition to information about residues involved in dictating the specificity of protein protein interactions. |
Exploitation Route | My findings will be used by researchers both within academia and industry. I was fortunate to collaborate with a team from Roche Pharmaceuticals who are keen to use the outcomes of our research to analyze large biomedical datasets. In addition I co-founded a small company that specializes in analyzing large, high dimensional datasets for commercial organizations in different contexts |
Sectors | Agriculture Food and Drink Digital/Communication/Information Technologies (including Software) Financial Services and Management Consultancy Manufacturing including Industrial Biotechology Pharmaceuticals and Medical Biotechnology |
Description | My findings will be used by researchers both within academia and industry. I was fortunate to collaborate with a team from Roche Pharmaceuticals who are keen to use the outcomes of our research to analyze large biomedical datasets. In addition I co-founded a small company that specializes in analyzing large, high dimensional datasets for commercial organizations in different contexts. Recently DeepMind research used the findings from this EPSRC funded research to build their alphafold protein structure prediction platform that dominated the 2018 CASP protein structure prediction contest. |
First Year Of Impact | 2013 |
Sector | Creative Economy,Digital/Communication/Information Technologies (including Software),Environment,Financial Services, and Management Consultancy,Manufacturing, including Industrial Biotechology,Pharmaceuticals and Medical Biotechnology |
Impact Types | Societal Economic |
Description | Marie Curie fellowship |
Amount | € 100,000 (EUR) |
Organisation | Marie Sklodowska-Curie Actions |
Sector | Charity/Non Profit |
Country | Global |
Start | 01/2015 |
End | 12/2018 |
Title | EVfold |
Description | Online server that allows users to predict tertiary protein structure |
Type Of Material | Improvements to research infrastructure |
Year Produced | 2012 |
Provided To Others? | Yes |
Impact | None yet noted |
URL | http://evfold.org/evfold-web/evfold.do |
Title | EVfold |
Description | Uses analysis of the correlation structure of large sets of protein sequences to predict tertiary protein structure from protein sequence data alone. |
Type Of Material | Computer model/algorithm |
Year Produced | 2011 |
Provided To Others? | Yes |
Impact | Widely used by other research groups |
URL | http://evfold.org/evfold-web/evfold.do |
Description | Predicting functionally important residues |
Organisation | Princeton University |
Country | United States |
Sector | Academic/University |
PI Contribution | Original research ideas, mathematical calculations and numerical simulations. |
Collaborator Contribution | Research ideas, funding of research assistants, funding of laboratory experiments to test the hypotheses generated by the calculations. |
Impact | Publication in genetics |
Start Year | 2012 |
Description | Predicting tertiary protein structure |
Organisation | Harvard University |
Department | Harvard Medical School |
Country | United States |
Sector | Academic/University |
PI Contribution | Original research ideas, algorithms and expertise. |
Collaborator Contribution | Provided funding to enable other members to join the research team and to supply computational equipment and working facilities. |
Impact | Publications in PLoS ONE and Cell. |
Start Year | 2011 |
Description | RMT |
Organisation | Harvard University |
Country | United States |
Sector | Academic/University |
PI Contribution | Original research ideas, mathematical calculations and numerical simulations. |
Collaborator Contribution | Original research ideas, mathematical calculations and numerical simulations. The collaboration also received additional funding via a grant made by roche pharmaceuticals to Harvard University. |
Impact | Publication in PRX. |
Start Year | 2011 |
Company Name | Vaticle |
Description | Vaticle operates a hyper-relational database for artificial intelligence systems, which enables businesses to manage their AI datasets in ways that traditional databases cannot. |
Year Established | 2013 |
Impact | Not yet available |
Website | http://www.grakn.ai |
Company Name | Aptamex Limited |
Description | |
Year Established | 2014 |
Impact | Not yet available |