University of Minnesota
University Relations

Minnesota Supercomputing Institute

Log out of MyMSI

Research Abstracts Online
January 2008 - March 2009

University of Minnesota Twin Cities
Medical School
Department of Neuroscience

PI: Tongbin Li, Associate Fellow

Support Vector Regression Approach to Capturing Peptide Sequence Characteristics; Developing Improved miRNA Target Prediction Tools

These researchers are using MSI resources for two projects. The first investigates peptide sequence characteristics. The primary structure (sequence) determines crucial properties of short peptides such as their binding affinities with histocompatibility complex molecules, and some categories of enzymes, such as protein kinases, phosphotases, and proteases. How to quantitatively represent features of a peptide pattern required for these reactions is a challenging problem. These researchers are using the support vector regression method to attack this problem, because of its demonstrated superiority in generalization/prediction performance among available machine learning tools. Preliminary exploration using a set of peptide array data suggests that a good performance can be achieved with reasonably small sizes of peptide binding data.

The second project, begun during this period, investigates microRNAs (miRNAs), a class of newly discovered genes capable of post-transcriptionally regulating the expression of other genes (their "targets”) by binding to the non-coding regions of those genes, leading to cleavage of transcripts and/or repression of translation. Despite efforts by many research groups over the past few years, the mechanism of miRNA targeting remains elusive. These researchers are collaborating with two groups at Stanford University in California, who have provided them with two sets of AGO2 immunoprecipitation microarray datasets and two sets of gene expression microarray datasets. These high-quality datasets offer a unique opportunity for elucidating the miRNA targeting mechanism. The group is applying a strategy where configurations for a dynamic programming-based scoring scheme are randomly created and summarized with supervised and unsupervised machine learning-based analysis methods, in order to achieve accurate miRNA targeting criteria with much improved coverage over existing methods.

Group Members

Gang Chen, Visiting Researcher
Wuming Gong, Research Associate
Xue Gong, Visiting Researcher
Shuli Kang, Visiting Researcher
Yongliang Ren, Research Associate
Luhui Weng, Visiting Researcher
Rendong Yang, Visiting Researcher
Liangsheng Zhang, Visiting Researcher