Top Banner
Bioinformati cs Bioengineering Summer Camp June 2001 Yaron Turpaz, Andrew Binkowski, Jie Liang
22
Welcome message from author
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Page 1: Bioinformatics

BioinformaticsBioengineering Summer Camp

June 2001Yaron Turpaz, Andrew Binkowski, Jie Liang

Page 2: Bioinformatics

• Introduction

– Basic molecular biology.

• What is Bioinformatics

• UIC’s Bioinformatics group

– Research projects

– Summer camp research project.

Outline

http://gila.engr.uic.edu/bioinformatics/

Page 3: Bioinformatics

DNA

Protein

Nucleotides sequence

Gene expression = Protein production

Page 4: Bioinformatics

Exons & Introns

STOP F R L

Page 5: Bioinformatics

Biological Diversity

Bacteria

Fruit Fly

Human

Yeast

Page 6: Bioinformatics

The Genetic Code• From DNA sequence to Protein sequence.

• 20 Building blocks of a protein = 20 Amino acids.

Page 7: Bioinformatics

From protein sequence to Structure

APRKFFVGGNWKMNGDKKSLGELIHTLNGAKLSADTEVVCGAPSIYLDFARQKLDAKIGVAAQNCYKVPKGAFTGEISPAMIKDIGAAWVILGHSERRHVFGESDELIGQKVAHALAEGLGVIACIGEKLDEREAGITEKVVFEQTKAIADNVKDWSKVVLAYEPVWAIGTGKTATPQQAQEVHEKLRGWLKSHVSDAVAQSTRIIYGGSVTGGNCKELASQHDVDGFLVGGASLKPEFVDIINAKH

=

Page 8: Bioinformatics

Bioinformatics

http://gila.engr.uic.edu/bioinformatics/

• Computational analysis of high-throughput biological data – Whole genome sequencing.– Global genomic expression & profiling.– Functional genomics.– Structural genomics/proteomics– Comparative genomics.

Page 9: Bioinformatics

Bioinformatics in BioE

http://gila.engr.uic.edu/bioinformatics/

• Interdisciplinary approach– Computer science, Mathematics & Statistics.– Molecular biology, Biochemistry & Medicine.

• Rapid growing impact area of BioE:– Boston U, UCBerkeley, UCSD, Rice, WashU, MIT,..

Page 10: Bioinformatics

Research

• Structural genomics/proteomics– Structural basis of functional motifs in protein families.– The CAST server - http://cast.engr.uic.edu/

– Drug discovery.

• Functional genomics– Collaboration with TIGR - http://www.tigr.org/

• Data mining of microbial DNA sequences for detection of foreign DNA.

• Whole genome comparative studies.

• Gene expression analysis– Collaboration with cancer biologist (Dr. Westbrook, School of Medicine)

• Molecular based informatics method to facilitate diagnosis of cancer.

Page 11: Bioinformatics

Gene expressionGene expression

Computational analysis of cDNA microarray expression profiles

~26,000 genes in one experiment

Page 12: Bioinformatics

Functional correlation in clusters

1. Cell division

2. cell signaling/cell communication

3. cell structure/motility

4. cell/organism defense

5. gene/protein expression

6. metabolism

7. unclassified

•Genes with similar expression patterns may participate in the same pathway or may be co-regulated.

•Clustering of expression patterns may reveal such relationships.

Page 13: Bioinformatics

Research – structure & function

How Proteins Interfaces with Other Molecules

• Analysis of protein topographic surfaces: – Identify protein function.– Predicting binding specificity and affinity.– Discovery of functional similarity.

• Protein interaction with cosolvents

– Stabilization of protein solution for longer

shelf life.– Molecular mechanism and optimization.

Page 14: Bioinformatics

Structural genomics/proteomics

• Atlas of Topographic Surfaces of All Known Protein Structures – Automatic identification of binding pockets.– Measurement size of surface binding pockets.

• Drug Discovery– Quantifying ligand accessibility. – Constructing precise negative imprint or cast of binding site.

Page 15: Bioinformatics

Geometry based approach for functional motifs

Discrete FlowDiscrete Flow

Voronoi Diagram and Delaunay TriangulationVoronoi Diagram and Delaunay Triangulation

Page 16: Bioinformatics

3D alpha shapes (HIV-1 protease)

Page 17: Bioinformatics

What is an algorithm?

• Precisely defined procedure for accomplishing a task.– driving directions,

– furniture assembly instructions,

– computer programs.• Built in hardware: fast• Built in software

Page 18: Bioinformatics

Are computers fast enough?

• NP-complete problems:eg. Traveling salesman problem:

20 cities a few seconds

30 cities a few hours

60 cities a few decades

• Computer speed does not increase exponentially.

Page 19: Bioinformatics

Pockets in Ribonuclease APockets in Ribonuclease A

Page 20: Bioinformatics

A Server for Identification of Protein Pockets & Cavities

CASTp

http://cast.engr.uic.edu/

Users of CASTp:Columbia, Harvard, Mayo Clinic, Princeton, Stanford, U Penn, SUNY Stony Brook, Texas A&M, UCIrvine, UBC, Virginia Tech, Yale, Abbott Lab, Pfizer, SmithKline Beecham, ...Agouron, Emisphere, Vertex, ...Kyoto U, Cambridge, European Molecular Biology Lab, INRA (France), Pasteur-Lille, Uppsala, Weizman,...Brazil, Czech, Korea, Turkey, ...

Page 21: Bioinformatics

CASTp ResultsCASTp Results

Calculations:

• Identifies all pockets and cavities. 

• Measures the volume and area analytically. 

• The number, area, and circumcircles of the mouth openings for each pocket. 

Files via email:

• pocket and mouth information file,

• Pocket and mouth atoms,

• a script file for visualization using rasmol.  

http://cast.engr.uic.edu/

Page 22: Bioinformatics

THANK YOU