Frederick NCI/MRMC/NIAID Combined Bioinformatics and Chemoinformatics Forum

2004 Projects for Master's Students

Quantifying Cell Motility

Cell motility and chemoattraction are vital processes that facilitate metastasis.  Currently, we are using time-lapse microscopy to image the movement of live cells.  Analysis software needs to be developed to calculate the speed of cell motion, and the degree to which the motion is random or directed.  Directed motion refers to motion towards a chemoattractant, or motion of cells towards (or away) from each other.  Experiments will be designed to compare this microscopy approach for measuring cell motility with the well-established Boyden chamber method to measure cell invasion.  The software will be used to study alterations in cell motility caused by candidate anti-cancer drugs and by altered levels of expression of cancer-associated genes. Contact: slockett@mail.ncifcrf.gov

Build an image databasing system

Currently, we have a microscope image archiving system with 600 Gigabytes of images, but no system to organize these images and mine them for information.  An intelligent image database system would enable searching of the archive for images with common properties.  Initially these properties would be keywords entered at the time the images were acquired.  Later developments would enable images to be computationally analyzed (ie segmented and measured) in order to extract quantitative measurements.  (We have considerable expertise in quantitative analysis and have a separate ongoing effort in that field.)  Also in the future, we will link into microscope image databases at other institutions.  We plan to build on the existing "open microscopy environment", which is supported by a consortium of scientists at NIH, Boston and Scotland . Contact: slockett@mail.ncifcrf.gov

Database and molecular evolution of ATP-binding cassette (ABC) transporters

Our lab studies ABC transporters and their role in human disease as well as the evolution of the genes in eukaryotes and the function of the genes in model organisms.  ABC genes are involved in a wide array of human genetic diseases (cystic fibrosis, adrenoleukodystrophy, cholesterol transport defects, and retinal degeneration.  In addition the genes are important in the efflux of drugs from cells and in multidrug resistance.  The goals of the projects would be to develop a web-based, publicly accessible database of ABC genes and sequences, extend the annotation of new ABC genes in recently sequences organisms, and to perform evolutionary analysis of these sequences.  Individual Masters level projects could include:

•  Build a publicly accessible database of ABC genes, sequences, alignments, expression data and links to other resources.

•  Annotate ABC genes in a newly sequenced organism and compare the sequences to those of other species.

•  Analyze the evolution of one of the 8 subfamilies of ABC genes, including alignment, phylogenetic analysis, gene loss and duplication and intron position.

•  Categorize and analyze the spectrum of mutations in human genetic disease for one or several related ABC genes.  Correlate mutation position with protein domain and predicted location in the 3D structure. Contact:

dean@ncifcrf.gov

Help build an intelligent genetic counseling system

We are building a new computer program, GENINFER, to assist genetic counselors in evaluating the risk of recurrence of genetic disorders based on the analysis of family pedigrees. The present version of the program uses a Java applet to provide a convenient graphical interface that permits counselors to draw, examine and modify family pedigrees and to enter information relevant to risk analysis. It also includes a general-purpose Bayesian inference mechanism that permits the rapid calculation and display of probabilities of various genotypes, for the consultand and all other pedigree members. This is possible even in the presence of complex pedigrees with multiple consanguineous matings. The ability to support rapid calculation also enables the user to perform sensitivity analyses. Current limitations of the prototype include a restriction to single-locus Mendelian disorders and an inability to make use of phase information. Planned extensions include remedying these limitations, the incorporation of an algorithm for automated reformatting (layout) of an existing pedigree, improvements in the population genetic models used by the program, and connections to external databases for acquiring data on disease incidence, patterns of inheritance, mutation rates, penetrance, etc. Development is in collaboration with the Clinical Decision Making Group at the MIT Laboratory for Computer Science. Contact: goldb@ncifcrf.gov

Recoding and Enhancement of the Expression Analysis Systematic Explorer (EASE) Software Application

EASE is a highly flexibly, standalone software application designed to facilitate the biological interpretation of genome-scale data sets such as those derived from gene expression microarrays and proteomic platforms. EASE identifies salient biological phenomenon through a statistically rigorous determination of over-represented functional categories within a given list of genes. Although EASE has addressed several key bioinformatic challenges, important enhancements are required in order to continue to meet the growing challenges posed by functional genomic studies. These enhancements include: 1) batch processing of experimental data, 2) summarized data visualizations, 3) drill-down and exploratory functions, 4) comparative analysis of multiple data sets, 5) and several others. The current implementation of EASE, including its graphical user interface, is written entirely in Perl. The main purpose of this project is to design an object-oriented version of EASE written in Java that includes the aforementioned enhancements.

The participating student will be responsible for object-oriented design and analysis of the enhanced software application and expected to present results at laboratory meetings and one scientific conference. The student will benefit from broad exposure and training in cutting-edge functional genomics technologies and bioinformatics concepts and will have potential for future employment in a leading information technology corporation.

Contact: gdennis@niaid.nih.gov

Develop computational approaches to RNA structure analysis

Our group studies the biological concepts of RNA structure function relationships and develops computational methodologies and tools to unravel these relationships. We have developed algorithms for RNA folding and analysis of the folding results. We were the first to develop a massively parallel (1000's of processors) genetic algorithm

(GA) for searching the very large RNA conformational space. In addition, we have developed a unique RNA structure analysis workbench, STRUCTURELAB, for analyzing the results of the GA and other folding algorithms. These tools, for example, have been used to study HIV and to determine RNA functional intermediates associated with RNA folding pathways. In addition, We have studied the three-dimensional behavior of RNA and RNA/Protein complexes using molecular dynamics and elastic

network interpolation. These computationally intense problems work with all atom and/or reduced atomic representations on state-of-the-art high performance parallel computers.

There are several possible projects that a student may work on related to computational approaches to RNA structure analysis. Some of the possibilities are listed below:

•  Develop WEB (Java) code for the RNA structure analysis system (STRUCTURELAB) that we have developed in our laboratory.

•  Develop computer algorithms for improving RNA structure prediction and analysis for both secondary and tertiary structure. This includes two-dimensional and three-dimensional modeling.

•  Develop data mining algorithms (e.g. data visualization and/or statistically based techniques) to interpret our genetic algorithm RNA folding results.

•  Structural searches for interesting RNA features present in RNA related biological systems using our software and other relevant software.

•  Further the understanding of RNA folding pathways using our genetic algorithm and the RNA structure analysis workbench STRUCTURELAB.

•  Design and implement an RNA structural database that will tie in with our software systems.

•  Carry out molecular mechanics, molecular dynamics, elastic network simulations using the supercomputer facilities for studying structural aspects of the RNA and RNA/protein interactions.

See also:

http://www-lecb.ncifcrf.gov/~bshapiro/index.html

http://www-lecb.ncifcrf.gov/~bshapiro/RNAstructure.html

Contact: bshapiro@ncifcrf.gov