Monday, 21 April 2008

Predicting RNA binding from protein sequences

The first Birkbeck seminar of the new term was given by Sue Jones, from the University of Sussex. Sue is no stranger to Birkbeck as she did her Ph.D. with Janet Thornton at University College, and later worked with her at the European Bioinformatics Institute and the biotech company Inpharmatica. Today she described a piece of software that she and her colleagues have developed for predicting motifs in protein sequences that are likely to bind to RNA.

Proteins function largely by interacting with other molecules - they are "social" molecules. Protein interaction partners include other proteins, carbohydrates, "small" molecules and ions, and the focus of today's talk: nucleic acids. The structures and functions of RNA molecules are diverse and include protein coding (mRNA), protein synthesis (tRNA and ribosomal RNA), splicing, hydrolysis of nucleic acid bonds (in RNA enzymes or "ribozymes") and control of gene expression (the so-called "micro-RNAs or miRNAs). RNA-binding domains in proteins include RNP domains, dSRNA binding domains, and K homology (KH) domains - all these are mixed (alpha and beta) structures.

Jones and her colleagues surveyed known structures of protein-RNA complexes and marked residues that were in close contact (through van der Waals or hydrogen bonding) with the RNA. They described each amino acid in terms of predicted accessible surface area, conservation within the family of homologous proteins, and chemical properties. Not surprisingly, positively charged and polar amino acids were favoured in binding to the negatively charged nucleic acid over negatively charged and hydrophobic ones; glycine, which is flexible, and tryptophan, which can form base stacking interactions were also favoured.

Jones then built these features, averaged over a "window" of 5-25 amino acids, into a support vector machine to predict RNA binding features in proteins of unknown function. (This technique is a form of "machine learning"; you don't need to know about it for this course, but if you're interested in knowing more and can cope with maths at a relatively high level, see the Wikipedia entry.) This was found to be at least as reliable as any similar tools that are publicly available.

There will be more about protein-nucleic acid binding in the next section of course material, Protein Interactions and Function, which is due to be released at the end of April.

Monday, 14 April 2008

British Crystallographic Association Spring Meeting

I spent Tuesday - Thursday of last week at the British Crystallographic Association Spring Meeting. The meeting has 4 strands Biological, Chemical, Industrial and Physical Crystallography.
Each contribute a plenary and then have their own separate sessions. The Biological Plenary was the Bragg Lecture where one famous crystallographer speaks about another usually older crystallographer and their work. This year Tony Crowther from the MRC Cambridge talked about his work and that of Michael Rossmann from Purdue University. Both made seminal contributions to the method of molecular replacement in protein crystallography. More on that in TSMB. Michael Rossmann when a postdoc for Max Perutz at the LMB in Cambridge was the first person to realise that the chains of hemoglobin looked like the chain of myoglobin and hence that you could solve structures of related proteins by molecular replacement. Michael then developed the mathematics and early software for molecular replacement. Tony Crowther did a Ph.D. with David Blow at the LMB and developed an improved form of the translation function. While working on natural language processing inEdinburgh, Tony also realised how to give a much faster and more acccurate version of the rotation function, which was the basis of the molecular replacement method for a long time. Tony's career was actually mainly in electron microscopy, he returned to the LMB from Edinburgh to work for Aaron Klug and became a group leader in his own right. Both he and Michael Rossmann have done most of their work on viruses and he talked about Michael's work on bacteriophage and his work on Hepatitis.
The Biological Group sessions were on Membrane proteins. Chris Tate from LMB in Cambridge described work they have been doing to stabilise membrane proteins by mutation. They search for alanine mutations that increase the stability of the protein in detergent and then carry out mutations in combination until the protein is stable for half an hour at a temperature 15-20 degrees hotter than the original. They have succeeded in crystallising beta-1 androgenic receptor, which will give important comparison to the beta-2 published just before Christmas.
The most interesting talk for me was from Thomas Sorenson, now at Diamond, on the work he and colleagues had done in the group of Poul Nissen in Aarhus. The group have published structures in several states of eukaryotic ATPase transproters (Calcium, sodium, proton). Interestingly these proteins were discovered by a Dane, Jens Skou and the group used proteins provided by various groups in Denmark that have worked on the systems for many years. They used natural sources and did not purify the proteins down columns, but just used differential extraction. This means that they isolated the membrane fraction that contained most of the protein and then extracted with detergents and this material was pure enough to crystallise in the presence of the right combination of detergents and lipids. The other biological sessions were on neutron diffraction, probing fast biological reactions, complementary methods, and ligand binding and drug design. Neutron diffraction gives the position of hydrogen atoms as both hydrogen and even more so deuterium diffract neutrons much more relatively than they do X-rays and you get density for hydrogen atoms. The catch is that neutron fluxes are much weaker and you need crystals that are 0.1- 1 mm3 compared to 0.0001 mm3 for a protein crystal. Studying reactions in crystals often means trapping intermediates by freezing out. Arwen Pearson from Leeds gave a good talk about a redox system that she had worked on in Minnesota where the reaction cycle can be carried out in the crystal, even changing space group between states. The catch, and this is common, is that X-rays themselves generate free radicals which can reduce redox centres so by collecting the data the redox state is altered. This meant that they had to collect data from several crystals before they became too damaged. The highlight for me of the Ligand and Drug session was a talk from Chris Phillips at Pfizer about their new non-nucleoside HIV Reverse Transcriptase inhibitor. These target a hydrophobic pocket in the protein, and tend to be rather 'greasy'. The Pfizer group had carefully designed a ligand that was both smaller and more hydrophilic and hence a better drug in terms of bioavailability.
There were many more great talks, but I hope this gives you a flavour of the meeting

Greetings from Poznan

This is just by way of an apology for my relative silence on PPS blogs and forums lately.

I am half way through two weeks' teaching at Adam Mickiewicz University, Poznan, Poland, funded by a grant to Birkbeck through the EU Erasmus programme (formerly known as Socrates) which funds student and lecturer exchanges between EU countries (and some others). I am teaching a two-week course on bioinformatics mostly, this year, to postgraduate Physics students but I have also taught in other departments.

I have known my host here, Professor Mariusz Jaskolski, since we were both working in the same lab in the States, NCI Frederick, in the early 90's. Mariusz was involved in some of the early work on the structure of HIV protease which is covered extensively in the PPS course. Since then he has gone on to found the first X-ray crystallography group in central-eastern Europe and to solve the structures of viral integrases, asparaginases and others... and many of his students, and others at AMU, have taken PPS and/or other distance learning courses from Birkbeck.

Normal service will be resumed next Monday.