BMI 731/IBGP 731 Biomedical Data Management

WINTER QUARTER 2004

(http://columbus.bmi.ohio-state.edu/~umit/bmi731/bmi731.html)

Umit Catalyurek, PhD, Department of Biomedical Informatics e-mail: catalyurek.1@osu.edu

Joel H. Saltz, M.D., Ph.D., Department of Biomedical Informatics

Ilya Ioshikhes, PhD, Department of Biomedical Informatics

Daniel Janies, PhD, Department of Biomedical Informatics

Catalin Barbacioru, PhD, Department of Biomedical Informatics

Tuesday & Thursday 9:00am – 10:30am

Location: 3167 Graves Hall

Tentative Reading Material: http://columbus.bmi.ohio-state.edu/~umit/bmi731/reading.html

Date	Lecturer	Topic
January 6, 2004	Daniel Janies	Biology
January 8, 2004	Catalin Barbacioru	Haplotype - 1
January 13, 2004	Catalin Barbacioru	Haplotype - 2
January 15, 2004	Daniel Janies	Sequence Databases Readings: Philips’00 and Wheeler’04
January 20, 2004	Ilya Ioshikes	Sequence Analysis Readings: [1] [2] [3] [4]
January 22, 2004	Daniel Janies	Tree-based Alignment
January 27, 2004	Daniel Janies	Phylogenetics
January 29, 2004	Ilya Ioshikes	Gene Regulation Readings: [1] [2] [3] [4]
February 3, 2004	Catalin Barbacioru	Microarray - 1
February 5, 2004	Catalin Barbacioru	Microarray - 2
February 10, 2004	Ilya Ioshikes	Mass Spectrometry
February 12, 2004	Joel Saltz	Translational Research
February 17, 2004	Ilya Ioshikes	RLGS
February 19, 2004	Joel Saltz	Radiology Imaging
February 24, 2004	Joel Saltz	DCE-MRI
February 26, 2004	Joel Saltz	Software systems: Data Warehouse
March 2, 2004	Umit Catalyurek	Multiple Sequence Alignment in Multi-Client Environment
March 4, 2004	Umit Catalyurek	Virtual Microscopy
March 9, 2004	Umit Catalyurek	Protein Structures and Related Database Searches
March 11, 2004	Umit Catalyurek	Sequence Compression [1] & Indexing and Filtering for Similarity Search [2]
March 16, 2004	Exam Week
March 18, 2004	Exam Week	Project presentations

Student Evaluation

Class participation 20%

Homework 30%

Term project 50%

Homework #1 - Due Tuesday, January 13^th

The table (HW1Table.xls) contains genotypic information about 4 (biallelic) SNPs (loci) in 927 samples. Missing data is denoted by NA (make sure you treat it appropriately)!!! The questions you have to answer are:

a) Are these SNPs in HW equilibrium?

b) Test linkage disequilibrium for AA haplotypes for all 6 possible combinations of 2 SNPs.

Homework #2 - Due Thursday, January 29^th

1. Find protein homologs to the protein query sequence (alpha-fetoprotein precursor from Homo Sapiens) using BLASTP program (http://www.ncbi.nlm.nih.gov/).

2. Find DNA homologs to the same query sequence using TBLASTN.

3. Perform a global pairwise sequence alignment of the protein query sequence from 1. with the top homolog found in 1. using GAP (SeqWeb program from http://gene.med.ohio-state.edu/gcg-bin/seqweb.cgi)

4. Find the segments of best similarity between top DNA homologs found in 2. using local optimal sequence alignment program BestFit of SeqWeb.

5. Make multiple sequence alignment of the top protein homologs found in 1. using progressive pairwise alignments (program PileUp of SeqWeb).

6. Find a consensus sequence based on results obtained in 5. using PRETTY program of SeqWeb.

7. Using programs for pair-wise and multiple sequence alignment and database searches, perform a comparative analysis of similarity of conserved segments in promoters and coding regions of genes, most similar to a gene encoding entire protein with a given fragment:

 >query

MAKNTAIGIDLGTTYSCVGVFQHGKVEIIANDQGNRTTPSYVAFTDTERLIGDAAKNQVALNPQNTVFDAKRLIGRKFGDAVVQSDMKHWPFQVVNDGDKPKVQVNYKGESRSFFPEEISSMVLTKMKEIAEAYLGHPVT

About GCG/SeqWeb:

Input Sequences: since you all use the same account, the sequences you added stay together, which is confusing. One way to solve this is, for each alignment operation, put all input sequences (in FASTA format, including header line) to be aligned into one txt file (with at least one empty line between each sequence), and then use Add From Local File to load the sequences. This should lead to the newly added sequences be highlighted in the box. Don't change or click anything in the box, and directly click Run to run the alignment. Also pay attention to the sequence type (DNA or peptide) and choose the right service in the menu.

What to submit: email to: ioschikhes-1@medctr.osu.edu

For each of the seven exercises:

1,2.: The top four sequences (better from different species) in the search result, in FASTA format. Better in one txt file.

3.-6.: The alignment result, better in html format.

7.: The alignment results, better in html format. And analysis report.

Homework #3 - Due Tuesday, March 9^th

Pick 7 sequences of 50S ribosomal protein L1 from different organisms using NCBI Web-site (http://www.ncbi.nlm.nih.gov). Perform multiple sequence alignment of the sequences using following programs: 3w (http://3w.molgen.mpg.de/3w.html), DCA (http://bibiserv.techfak.uni-bielefeld.de/dca/), ClustalW (http://www.ebi.ac.uk/clustalw/), and Dialign (http://bibiserv.techfak.uni-bielefeld.de/dialign/). Try to analyze the results obtained by each program and explain the differences.

Email the results by each program (preferably in HTML format) and analysis report (in Word format) to ioschikhes-1@medctr.osu.edu by March 9^th.

Sample project topics:

1. Image Analysis: Implement a software system that will automatically retrieve images from a PACS system (BMI will run a software PACS system on its own machines) and will execute an image analysis code, such as segmenting nuclei
2. DICOM/PACS: Design, architect and implement a picture archival system that will run on distributed heterogeneous cluster of machines.
3. Classification of tumors and identification of differentially expressed genes using gene expression data.
4. Search for new elements in promoters of human genes using sequence analysis.