An efficient combination of molecular biology and high performance computing makes possible a remarkably fast development of this new scientific area. Currently modem supercomputers and new algorithms are used to predict structures and properties of species consist of hundred of thousand of atoms or to obtain accurate information on molecules with more than 100 atoms. Computational chemistry has been successfully applied to reveal many aspects of biological systems.
Practical applications are obvious from the following example. Since a number of potential molecules to be considered in searching for a new drug has been estimated to be between 1060 and 10400, it is impossible to synthesize even a small fraction of these compounds. Computational chemistry can address two aspects of this problem--the calculation of many compounds in order to eliminate the unnecessary work from the laboratory regarding all inactive species but also the prediction (design) of compounds with the desired properties.
The basic interest is in describing properties of the DNA components. Among these are the nucleic acid bases (NAB). Their tautomeric equilibria might be responsible for the spontaneous mutations in DNA. Their hydrogen bonding and stacking interactions govern the structural stability and variability of the nucleic acid. DNA bases and their interactions are evaluated at the ab initio levels with inclusion of electron correlation. Such predictions are quite accurate and replace hard-to-obtain experimental data.
Fundamental properties of biological systems can be predicted based on the results of multiple analysis. Such an approach allows for the build-up of a library of different compounds and allows for the use of it for evaluating of new species.
Computational methods are capable of predicting the effects of an aqueous environment on the structures and properties of biomolecules. Of interest is the development of a fast and efficient algorithm to characterize the interactions of a solute with a polar solvent. Since ab initio calculations are still of limited application for big biomolecules, such a method has to take advantage of a semiempirical approach. Currently a number of salvation models have been tested on different compounds and their successes and limitations are of general interest.
Computer modeling studies can also reveal the details of multistrand structures of DNA sequences. In addition, the computational predictions of 3-dimensional structures are becoming more accurate and help to obtain predictive models of three dimensional protein structures. A very challenging problem--the prediction of protein structure from amino acid sequences--have been recently addressed. Closely related to DNA problems is the phenomenon of proton transfer. Such a process can be considered for the hydrogen bonded NAB or DNA bases interacting with a solvent (water) shell. It can be also described as a dynamic process and can give information about the time evolution of biological systems. Of interest are also new approaches to make such models computationally efficient.
Crucial for biological systems are the interactions of membranes with small compounds (drugs) and with peptides which have been studied by computational methods, and the results of such calculations reveal the details of such interactions. In addition, a new method allows for the prediction of charge-transfer in natural membranes.
The organizers of the Networking Resources for Collaborative Research in the Southeast Conference, held June 3-5 at the Georgia Institute of Technology in Atlanta, Georgia addressed the needs of the scientific community to establish interdisciplinary collaboration among researchers working in this field and they included a molecular biology session in the agenda of the conference.
The first session (June 3) consisted of three talks. Dr. Stanley Burt, from the Frederick Biomedical Supercomputing Center, spoke on new challenges in Computational Molecular Biology. He also discussed results of current studies at the Frederick Biomedical Supercomputing Center. Among the topics of his talk was an overview of using structures to design drugs.
The second speaker, Dr. Jiri Sponer, from the J. Heyrovsky Institute of Physical Chemistry, then addressed the issues of combining high levels of ab initio calculations on biomolecules with data obtained from experimental studies and molecular mechanics calculations. He discussed advantages of molecular dynamics approach in studies of biological systems. He concluded by discussing possible ways of improving existing force field parameters.
The last speaker of the session was Dr. Charles Liarakos. Dr. Liarakos, from Directorate for Biological Sciences, gave an overview of molecular biology areas that are currently supported by NSF. In conclusion, he also discussed new and emerging areas that have derived from these studies.
The second session (June 4) was a break-out session. This session attracted the largest pool of conference participants (30). During the breakout storming sessions, a number of topics were identified for possible collaboration: The topics were as follows:
The third session (June 5) involved discussions in smaller groups. From these sessions, five groups have been established. The topics covered by these groups are as follows:
Members of these groups developed plans for further collaboration efforts. At the end of the meeting the molecular biology group reported the developed action plans.
The molecular biology session has addressed important needs of researchers working in this area. Since the background of this group is quite diverse, (chemistry, biology, physics, computer science, mathematics), an organization of a forum for common discussion has facilitated interactions and communication among participants. The lectures presented by three invited speakers established a good starting point for such interactions. Most of the participants were very enthusiastic about their own projects and were able to find a common interest with other researchers. The discussed topics were diverse ranging from the establishment of the WEB-based database for students to sophisticated research on DNA fragments. The level of enthusiasm and developed plans assure that established collaborations will be continued after the meeting. A second meeting, a year after the current one, would be an efficient addition and follow-up of such plans.
Back to Conference Home Page