Statistical And Computational Methods For Molecular Biology And Biomedicine

Computer Research And Technology

Investigators

Linked publications & trials

Abstract

In a joint study with investigators in Laboratory of Molecular Biology, NCI and Institut National de la Recherche Agronomique (INRA), France, we are attacking the problem of protein structure classification, with the goal of improving automated methods for recognition and classification of protein domains in three dimensional structures. Domains are thought to be the building blocks of complex structures, and often determine protein function. We have recently shown that two distinct structure similarity measures (VAST and SHEBA) can obtain at best about 75-80% agreement with a standard manually curated protein classification (SCOP), calling into question the existence of sharp boundaries between protein "folds". We have completed a further analysis of "C-class" or alpha/beta protein domains, by hierarchically clustering domains based on measured structural similariaty by three different methods (VAST, SHEBA and DALI). We explored novel dendrogram-cutting algorithms as a means of automatically generating novel classifications, and compared several of these to SCOP and find that automatic classifications differ little from each other, relative to their overall differences from SCOP. One implication is that identification of conserved motifs or cores may be a necessary step in identifying domain classes, and multiple as distinguished from pairwise structural alignment may be required to accomplish this.[unreadable] [unreadable] With an investigator in the Division of International Epidemiology and Population Studies, Fogarty International Center, we have developed a phenomenological model of plasmodium/red blood cell dynamics moderated by host immune and erythropoietic responses. We have found that the tight synchronization of the asexual reproductive cycle of the parasites seen clinically may actually be beneficial to host by limiting parasitemia and severity of anemia. We are now studying how regulation of parasitemia of both host and parasite factors affects transmissibility of the parasite to its mosquito vector, and to further develop our models, we are working with laboratory investigators in the Laboratory of Malaria and Vector Research, National Institute of Allergy and Infectious Diseases.[unreadable] [unreadable] With investigators in the Section on Medical Biophysics, Laboratory of Integrative Biophysics, NICHD, and with the Signal Processing and Instrumentation Section, Computational Bioscience and Engineering Laboratory, Division of Computational Bioscience, CIT, we are working of a model of the thermal and fluid transport processes that occur in the operation of expression microdissection (xMD), a newly developed method of extraction large number of cells from a tissue sample. xMD shows great promise, but a better theoretical understanding of its operations is needed before it can commercially developed or be fully exploited in an NIH core facility.[unreadable] [unreadable] With investigators in NIMH, we analyzed multiple-electrode recordings from in-vitro neural network preparations in order to deduce the underlying cortical network topology. We found that such functional networks show a strong "small world" property, meaning high clustering among the nodes and short node-to-node distances. More importantly, we found a novel property of these networks when the weights of the network links are taken into account. Simulations indicate that such network architecture results when the link weights are correlated with the clustering coefficients of the respective end nodes. [unreadable] [unreadable] With investigators in the Laboratory of Integrative and Medical Biophysics, NICHD related to the development of the optical imaging techniques, we derived theoretical predictions for the effects of the photon-fluorophor interactions in time-gated optical imaging techniques. In another project with LIMB, NICHD, related to the development of the diffusion tensor MRI, we use tensor-normal distributions to derive the statistical distribution of the uncertainty in the orientations of the neural fibers which can be measured with DT-MRI.