Ph.D. Training in Computational Biology

Areas of Research

Faculty Members

Ziv Bar-Joseph
Dr. Bar-Joseph's primary research is in computational and systems biology. Machine learning and statistical algorithms are used to study time series expression experiments and to combine high throughput biological data sources to model biological networks.

Dannie Durand
The Durand group works in comparative genomics, focussing on the evolution of genome organization and functional diversity in vertebrates.

Christopher Langmead
The Langmead laboratory's interests encompass the areas of structural and temporal phenomena in biological systems. Projects range from new techniques for problems in structural biology to modeling of the 3D growth of brain tumors

Jonathan Minden
A major objective in the Minden laboratory is developing new tools for comparative proteomics. An important aspect of this proteomics work is to develop automated image analysis software to detect and rank protein changes during development and disease.

Robert F. Murphy
Computational cell biology projects in the Murphy laboratory include using computational approaches to analyze the patterns of proteins in fluorescence microscope images to create self-populating knowledge bases and models of protein subcellular location.

Gordon S. Rule
Computational interests in the Rule laboratory include developing methods for the rapid assignment of NMR spectral peaks and, in collaboration with Dr. Michael Erdmann of the Computer Science Department, developing geometric methods to detect structural homologies in proteins from NMR data.

Russell Schwartz
The development of modeling and simulation techniques for studying self-assembly systems, as well as methods for analyzing and applying genome polymorphism data are a primary area of interest in the Schwartz laboratory.

Nathan N. Urban
In the Urban laboratory researchers develop highly reduced and highly detailed models of single neurons and neuronal circuits to understand how the properties of these cells and circuits contribute to the generation of phenomena such as oscillations, synchrony and other complex dynamical behaviors.

Eric Xing
Eric Xing developes statistical models and machine learning algorithms for biological network inference and characterization, cis-regulatory module decoding, regulatory evolution modeling, quantitative trait locus mapping, genome polymorphism patterning, and population genetic analysis. He is also applying these quantitative approaches to investigate the mechanisms of breast cancer development and metazoan morphagenesis.

Graduate Student Work in Computational Biology

Amina Salim Abdullah
Yanhua Hu
Juchang Hua

Arun Krishnaswamy
Arun works on applying methods from machine learning to the problem of predicting nuclear magnetic resonance (NMR) chemical shifts from protein structure. Chemical shifts are among the most precisely measurable NMR spectral parameters, and a better understanding of the relationship between shift and structure will significantly advance experimental and computational protein structure determination methods.

In this effort, a specific family of machine learning algorithms termed ensemble methods is employed in conjunction with a detailed description of the protein structure of interest in terms of numerical and nominal features to predict shifts. The objective is to have the machine learning method learn a mapping between the structural description and the experimentally determined chemical shifts, and use this learned model to predict shifts for novel structures. An accurate shift prediction protocol thus developed has applications in refining existing protein structures, in the difficult task of NMR shift assignment and in problems such as decoy detection.

Narayanan Raghupathy
Bormi Shin
Nan Song
Tiequan Zhang

Recent Publications and Presentations

Durand D, Halldorsson BV, Vernot B. A Hybrid Micro-Macroevolutionary Approach to Gene Tree Reconstruction. J Comput Biol. 2006 Mar;13(2):320-35.

Przytycka T, Davis GB, Song N, Durand D. Graph Theoretical Insights into Evolution of Multidomain Proteins. J Comput Biol. 2006 Mar;13(2):351-63.

Durand D, Hoberman R. Diagnosing duplications--can it be done? Trends Genet. 2006 Mar;22(3):156-64.

QuickLinks for Computational Biology

Computational Molecular Biology Symposium

Joint Carnegie Mellon - University of Pittsburgh Ph.D. Program in Computational Biology