Sacan Bioinformatics Lab

Sacan Bioinformatics Lab is part of the Center for Integrated Bioinformatics at the Drexel University and is led by Ahmet Sacan. Our main research interests are:

  • Structural Bioinformatics
    • Searching and mining protein structural data
    • Pairwise and multiple structure alignment
    • Secondary Structure prediction
    • Subcellular localization
    • Remote homology detection
  • Expression analysis
    • Microarray experimental design
    • Gene set enrichment analysis
  • Biomedical Imaging
    • Automated cell tracking
    • Modeling and analysis of cell locomotion

The following is a list of our software and web-services:

  • Vorometric: Integrated search and alignment of protein structures using Voronoi contacts
  • Smolign: Spatial Motifs Based Multiple Protein Structure Alignment
  • CellTrack: An Open-Source Software for Cell Tracking and Motility Analysis
  • MicroarrayDesigner: An Online Search Tool and Repository for Near-Optimal Microarray Experimental Designs
  • LFM-Pro: A Tool for Detecting Significant Local Structural Sites in Proteins

Research Projects

Identification of Protein Functional Sites from Novel Surface Representations

Protein Structure Initiatives are generating experimentally determined structures of proteins, and most of these target proteins have unknown functions. Annotation of these new protein structures requires careful comparison with the compendium of known functions. We are utilizing a novel representation of protein surfaces that map important biochemical features to 2D image representations that are computationally more practical to study. Comparison of these surface representations will aid in functional annotation and yield insights into functional evolution of protein families. The same surface representation will provide a means by which protein-protein interactions can be studied and predicted.

Gene Regulatory Network Reconstruction from Microarray Data:

Microarray experiments provide the expression levels of tens of thousands of genes and are nowadays routinely used to study cellular responses to stimuli. The ability to identify differentially expressed genes from microarray experiments has enabled discovery of genetic biomarkers for human diseases. In order to understand the biological mechanism of the identified genes, they are considered in the context of regulatory and metabolic networks. Our knowledge of these networks is limited to a small set of well-studied interactions. We are utilizing various genomic and expression data to reconstruct these networks. The integration of time-series microarray data, co-expression profiles obtained from different microarray datasets, sequence and evolutionary profiles of genes will improve the accuracy and coverage of the predicted interactions.

Micro-RNA Modulation of Gene Expression.

Micro-RNAs are a recently discovered class of small non-coding RNA molecules that degrade or inhibit translation of mRNAs. Aberrant Micro-RNA expression has been indicated in a variety of human diseases. Micro-RNA's regulate protein levels by destabilizing mRNA molecules and/or by their inhibiting translation. While several hundreds of conserved micro-RNAs have been identified, their effects on the gene expression levels are not well established. We are developing novel analysis methods for prediction of regulatory effects of micro-RNAs. Our case studies of microRNAs include their involvement in Chronic Regional Pain Syndrome and Spinal Chord Injury.

Integration of Biological Knowledgebases:

With the advent of high-throughput experimental methods in biology, and similar advances in information analysis methods has prompted many national and international efforts to catalogue the data generated from these high-throughput experiments. The heterogeneity of these data and differences in the data access methods have made an integrated utilization of the available data a major technical challenge. We are building novel data representation and communication methods to integrate the world of biological data. A graphical user interface utilizing this integrated data is being developed to enable its utilization for bioinformatics education and by non-programmers.

Learning to See: Segmentation and Tracking using Pre-annotated Images:

Many of the image processing and computer vision methods have been developed as general purpose methods whose application to domain-specific problems is non-trivial and labor intensive. We are building image segmentation and object tracking methods that can make use of a small set of pre-annotated images. A novel representation of existing information is utilized to train pattern classifiers that can segment new images or track new objects. The developed methods are being applied to wound images and live cell microscopy images.

Online Technologies for Programming Education:

Computer programming has become a required skill in engineering and is becoming so in many other data oriented fields. In response to this change, computer programming is being taught more widely in secondary and post-secondary education. Online technologies to assist in self-centered learning and evaluation have not been sufficient to provide for the expanded interest in programming. We are building ProgramminBank as an online technology to assist programming education in engineering and other fields. ProgrammingBank will host a repository of questions that are closely tied to learning objectives and will provide real-time assessment of student work.

index.txt · Last modified: 2011/11/29 09:18 by Ahmet Sacan