Bioinformatics of regulation and structure of genomes in Siberian

Science on the verge of artScience in Siberia No. 28-29, 22.07.2010

At the end of June, the Institute of Cytology and Genetics SB RAS hosted an International Conference on Bioinformatics of Genome Regulation and Structure (BGRS\SB'10). This is the seventh conference dedicated to this topic (such events are held every two years). The organizer and permanent Chairman of the Program Committee is the Director of the ICIG SB RAS, Academician Nikolai Alexandrovich Kolchanov.

The program included plenary lectures, oral presentations, specialized poster sessions, symposiums, computer demonstrations and software demonstrations. The sections discussed the issues of bioengineering of macromolecules, integration of in siliko and in vitro technologies in the development of medicines, current problems of systems biology and gerontology, etc. The symposiums were devoted to modern concepts in scientific research of laboratory animals, systems biology in parasitology and other interesting topics. Also, within the framework of "BGRS\SB'10", a scientific and practical conference "Day of Supercomputing Technologies: Science, education, industry" was held under the auspices of the Government of the Novosibirsk Region and the Council for Supercomputing at the Presidium of the SB RAS in partnership with Intel Corporation, as well as the School of Young Scientists "Bioinformatics and Systems Biology".

According to the scientists themselves, in terms of the versatility and breadth of the issues discussed, the BGRS conferences are perhaps the only events of this profile in Russia and the CIS countries. A distinctive feature of the current conference was the inclusion of the issues of systems biology in the range of discussed problems – a new direction whose task is to study the patterns of organization and functioning of biological systems of different hierarchical levels and the integration of the data obtained.

Our correspondent Elizaveta Sadykova met with the forum participants and learned their opinion both about the event itself and about trends in the development of modern biology.

According to academician N. A. Kolchanov, systems biology is a new area of research that has a pronounced integrative character, based on a combination of experiment and bioinformatics. The need for such integration arose due to the fact that in the last ten years there have been significant changes in experimental methods for studying living beings at all levels of hierarchical organization, from genomic and genetic, to the level of the organism as a whole.

A whole "family" of sequencing methods has appeared, allowing the whole genome to be read in one experiment. Today, scientists have in their arsenal high–performance methods for studying the proteome - the protein composition of cells, tissues, organisms; more advanced technologies for studying gene expression, high-performance and high-resolution tomographic analysis technologies that make it possible to study the structural organization of organisms in the lifetime and the distribution of various metabolites with good detail. The volumes of experimental data obtained on the basis of these technologies are so large that not only their practical use, but even their integration and interpretation are impossible without the use of modern information technologies, high-performance computer systems, effective data analysis methods and approaches that allow modeling the studied systems.

But it is impossible to survive alone in the modern world, so scientists combine their efforts, for which such conferences are held. Thus, the most reliable partners of ICIG are, of course, the institutes of SB RAS and NSU. Cooperation has also been established with French colleagues from INRA, German colleagues from Heidelberg (joint study of hepatitis C virus), scientists from India (study of tuberculosis). And, of course, the most sensational project is the sequencing of the opisthorchus genome, carried out jointly with specialists from scientific institutions in different cities and countries, including the Russian Scientific Center "Kurchatov Institute", formerly the I. V. Kurchatov Institute of Atomic Energy, Siberian State Medical University (Tomsk) and a major scientific center in Thailand.

Professor Alexey Sergeevich Ivanov, V. N. Orekhovich Research Institute of Biomedical Chemistry of the Russian Academy of Medical Sciences, told us about research in the field of biomedicine and computer modeling:

– Initially, I was an experimental biophysicist, my PhD was in biophysics of biological membranes, but gradually my scientific interests shifted towards biochemistry, in which I defended my doctoral degree. When Academician Alexander Ivanovich Archakov headed the Institute of Biomedical Chemistry of the Russian Academy of Medical Sciences (IBMH named after V. N. Orekhovich of the Russian Academy of Medical Sciences), he invited me to do molecular computer modeling. Then it was a novelty – even the word "bioinformatics" did not exist yet. And, nevertheless, we have engaged in an extremely difficult direction – computer design of medicines.

The laboratory slowly grew, progress did not stand still, more powerful computers appeared, "advanced" software, new discoveries in molecular biology and genomics expanded our knowledge, so a second computer laboratory was soon created at the institute, then a third, and as a result a whole department of bioinformatics appeared. We managed to make a breakthrough in the field of computer-aided drug design, so foreign pharmaceutical companies began to apply to the institute with proposals for joint research. So, for five years we have been carrying out a scientific project with Procter & Gamble, for seven years we have worked with the scientific center of Janssen-Cilag, and had contacts with other foreign companies. Thus, we earned extra–budgetary money, which was used to purchase the first professional computers and software for molecular modeling - powerful multiprocessor servers and working graphics stations from Silicon Graphics.

Academician A. I. Archakov managed to prove both at the academic and at the state level that computer modeling issues need to be dealt with seriously, that this is not a toy, but a necessary new component of the drug creation process. It does not cancel all classical pharmacology, but facilitates the first steps of the birth of a new drug.

– Alexey Sergeevich, did you manage to develop any medicine and bring it to the pharmacy?

– Usually, the work carried out within the framework of grants, research or government orders ends with the emergence of new prototypes of medicines. The further path of creating a real drug based on a prototype, the implementation of all regulated preclinical and clinical trials, in short, all stages of bringing the drug to production and pharmacy can take from 15 to 25 years and require a completely different level of funding. But we did make one medicine, and it is on sale - this is "Phosphogliv". The capsule form of our medicine is a more effective and safe analogue of Essentiale Forte in its properties, however, our main achievement, which was awarded the prize of the Government of the Russian Federation, is Phosphogliv for intravenous injections, created to restore damaged membranes of liver cells and has extremely high efficiency. This form of the drug is the first real nano-drug developed in Russia. It was made at a time when no one had ever heard of nanotechnology, and consists of nanoscale phospholipid particles. Since we were pioneers in this matter, industrial technologies of this level simply did not exist, we had to solve these problems ourselves, from scratch. For example, part of the production workshop was constructed at the Khrunichev NPO and represented half of the space station. As a result, a pilot production facility was built at our institute according to the GLP standard, which has been producing both forms of the drug for more than 10 years.

– Please tell us in more detail what the process of developing new medicines looks like?

– According to the classical scheme of pharmacology, new basic structures of future medicines are found in natural objects by a wide "blind" search, with the expenditure of huge funds and time and extremely low efficiency. The same thing happens in laboratories when chemists and pharmacologists try to modify the chemical structure of a known pharmacologically active substance in order to improve its properties. At the same time, it is almost impossible to find a fundamentally new compound that differs in structure from what is already known, researchers do not have any premises for this.

What allows you to do computer modeling? First, to determine the implicit relationships between a set of known compounds, their chemical and three-dimensional structure and the properties they exhibit. There is a whole direction in bioinformatics that deals with such analysis. Experiments, when the search is conducted blindly, lead to extremely irrational expenditure of effort, time, money, laboratory animals. But most of these attempts nowadays can be performed on a computer in a matter of minutes: a new chemical structure is drawn, and an expert system with a probability of 80-90% can give the correct answer about changing the properties of the modified compound. That is, the unproductive, looped experimental process of searching for successful modifications of certain compounds is taken over by the computer, reducing the amount of experimental work at this stage. But this in no way excludes all the steps of classical pharmacology, including preclinical and clinical trials of future drugs, but complements them with a rational approach in the initial search for new basic structures.

Relatively recently (about 15 years ago), the dogma of modern molecular pharmacology has developed: any drug compound has a pharmacological effect by interacting with its molecular target. Proteins are most often used as such targets. With the advent of this formulation, a new scientific direction has crystallized – the construction of drug molecules based on the spatial structure of the target protein. Science began to move rapidly forward in the development of methods for modeling the spatial structures of proteins, their complexes with low molecular weight ligands, assessing the strength of complexes by calculating changes in free energy, etc. However, all studies have come up against the problem of lack of knowledge about the three-dimensional structure of target proteins.

Due to genomic successes, we have huge knowledge about the amino acid sequence (the primary structure of a protein), we know literally millions of primary structures of proteins, but we have practically no knowledge about three-dimensional structures, only from 3 to 5% of this volume. Moreover, we have almost no idea about the function of proteins that are found in genomic studies. And the annotation of genomic data leaves much to be desired so far. The functions of proteins are annotated by the similarity of their amino acid sequences. If they are similar, then these are related proteins and their functions are similar. As a result, this function is attributed to each new protein without any verification. The result is entered into databases and can be used to annotate another protein, and the new protein will be annotated by a protein previously known. That is, hypotheses generate new hypotheses of the second, third and so on order. The number of errors at the same time grows from time to time. But while this method remains the only one, others have not yet been discovered. Today, bioinformatics is very actively developing various mathematical methods that allow you to annotate even proteins that are not similar to anything. So far, everything remains the same: protein annotations are just hypotheses that no one has experimentally tested.

There is another problem – many proteins are listed in databases, although no one has checked their physical existence in living organisms. They were calculated based on the analysis of genomic sequences as possible gene products. Therefore, it is unclear how new drugs can be designed for such hypothetical targets. A large area of bioinformatic, genomic and proteomic research is touched upon here, so at one time academician A.I. Archakov put forward a thesis about a triad of technologies in the post–genomic era - genomics, proteomics and bioinformatics, which cannot exist without each other.

A great success in modern protein science is the opportunity to obtain any proteins in large quantities by methods of genetic engineering, biotechnology and biochemistry. The approaches of classical biochemistry, when protein was isolated from the biological tissue of the studied living organism, are a thing of the past. Now there is no need to take whole organs or tissues to get almost any protein, for example, a rabbit, a mouse, and even more so a human. Modern methods are based on the creation of a gene construct containing a target protein gene inserted into a producing bacterium (for example, E. coli or yeast). If conditions are created for the reproduction of this microorganism, it is possible to obtain biomass and isolate the target protein from it in any necessary quantities. These technological advances have dramatically accelerated the production of target protein preparations in quantities necessary for the study of their spatial structure by protein crystallography methods.

– Are proteins able to crystallize?

– The protein crystal is a separate story. In order to study reliable information about the spatial structure of the target protein, there is practically the only way – to accumulate it in the right amount and grow a pure crystal. In this matter, science still remains on the verge of art. Growing a crystal is an art. Then everything is simple. The obtained crystals are subjected to X-ray diffraction analysis, as a result of which the researchers obtain the coordinates of the spatial arrangement of atoms in the target protein.

Since the crystallization of target proteins is very important for successful work on the creation of prototypes of new drugs, I hope that a comprehensive project on their crystallization in space, on the Russian segment of the ISS, will find its continuation. As part of this project, we performed ROC on the order of Rosskosmos. It was about the preparation of samples of target proteins and promising variants of prototype substances for their crystallization in conditions of cosmic weightlessness and the absence of thermal convection affecting crystal growth.

– What is your opinion about this conference, what are you particularly interested in here?

– This is not the first time the Conference has been held, and this year it has become especially noticeable that it is changing. In my opinion, this is good, science does not stand still, and it is clear that in our time there cannot be one "pure" bioinformatics without genomics, proteomics and a number of other related disciplines. Which automatically means that there can be no pure computer science without experimental science in the field of research of living systems. I do my best to promote the integration of these areas, because often the emphasis is only on bioinformatics.

New approaches in protein function annotation are interesting. Several reports have been made on this topic. Proteomics is a science that emerged after genomics, and its primary task is the inventory of proteins, that is, experimental confirmation of the presence of the final product (protein) for each gene. Until recently, it was believed that the human body is so complex, and there are so many genes in it, that there may be several million proteins, and it is not possible to engage in an end-to-end inventory of them all. But in 2000, the complete human genome was studied and documented. 9 years have passed, and the scientific community has thought about the Human Proteome project. An international proteomic community (by the type of genomic) was created. The fact is that, unlike the genome, the human proteome is more complex, because proteins can be different in different cells and in their different states. This is a very big job that the international community has decided to "swing at". The genocentric principle of dividing the sections of the work was adopted – by chromosomes. America chose 21 chromosomes, Russia - 18, and several other countries took one chromosome each for study. The task is to examine all the proteins of the target chromosome. Our institute has also joined this project.

Recently, it was decided at the level of the Government of the Russian Federation that these studies should be supported in every possible way and not spare money for them. Now two areas of work are being formed – Russia's participation in the international project "Human Proteome" (study of proteins of the 18th chromosome) and the national project of human proteome research. I think that these projects will become a point of integration of the efforts of various scientific institutions of different departmental affiliation. Undoubtedly, the institutes of SB RAS and SB RAMS will also make a great contribution.

Portal "Eternal youth" http://vechnayamolodost.ru23.07.2010