Top Banner
1 Computational Genomics Course Lecture 12 fall 2002/03 School of Computer Science Tel-Aviv University Instructor: Benny Chor Many slides taken (with permission…) from Metsada Pasmanik-Chor adiation Hybrid (RH) Mapping
23

1 Computational Genomics Course Lecture 12 fall 2002/03 School of Computer Science Tel-Aviv University Instructor: Benny Chor Many slides taken (with permission…)

Mar 29, 2015

Download

Documents

Cameron Nutting
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: 1 Computational Genomics Course Lecture 12 fall 2002/03 School of Computer Science Tel-Aviv University Instructor: Benny Chor Many slides taken (with permission…)

1

Computational Genomics CourseLecture 12 fall 2002/03

School of Computer ScienceTel-Aviv University

Instructor: Benny Chor

Many slides taken (with permission…) fromMetsada Pasmanik-Chor

Radiation Hybrid (RH) Mapping

Page 2: 1 Computational Genomics Course Lecture 12 fall 2002/03 School of Computer Science Tel-Aviv University Instructor: Benny Chor Many slides taken (with permission…)

2

A map - graphic representation that provides

information about the location of sites and the

spacing between them.

Maps for the genome provide the relative order of

items (“markers”) along the chromosome.

Genomic Mapping

Two Major Types of

Genomic Maps :• Genetic maps.• Physical maps.

1

Page 3: 1 Computational Genomics Course Lecture 12 fall 2002/03 School of Computer Science Tel-Aviv University Instructor: Benny Chor Many slides taken (with permission…)

3

Physical Maps

Cytogenetic (chromosome) map - based on the distinctive banding patterns observed by light microscope of stained chromosomes.

cDNA map - locations of expressed DNA (exon regions) on the chromosome.

Low resolutionLow resolution

High resolutionHigh resolution

Contig (cosmid) map - the order of overlapping DNA fragments spanning the genome.

Restriction map - describes the order and distance between DNA enzyme cleavage sites.

Sequence “map” - complete sequencing of a chromosome.

Radiation Hybrid (RH) map - the order of DNA markers (STS), each appearing uniquely in the genome.

2

Page 4: 1 Computational Genomics Course Lecture 12 fall 2002/03 School of Computer Science Tel-Aviv University Instructor: Benny Chor Many slides taken (with permission…)

4

Looking directly at chromosomes, using the technique of Fluorescence In-Situ Hybridization (FISH).

Metaphase chromosome spreads

For FISH demo look at:http://www.waisman.wisc.edu/cytogenetics/procedures/FISH/FISHmethod.html

Cytogenetic Map - A Low Resolution Map

3

Page 5: 1 Computational Genomics Course Lecture 12 fall 2002/03 School of Computer Science Tel-Aviv University Instructor: Benny Chor Many slides taken (with permission…)

5

Physical maps are a set of ordered DNA clones that cover the complete chromosome. These clones overlap each other to form a contiguous array (contig).

Physical Maps (Contig Maps)

contig assembly

clones

contig(contiguous array)

4

Page 6: 1 Computational Genomics Course Lecture 12 fall 2002/03 School of Computer Science Tel-Aviv University Instructor: Benny Chor Many slides taken (with permission…)

6http://opbs.okstate.edu/~melcher/MG/MGW1/MG11124.html

RFLP - Genetic Markers Creating Genetic Maps of Chromosomes.

A change of a single nucleotide may produce RFLP (restriction fragment length polymorphism), by changing locations of restriction enzyme recognition site.

5

Page 7: 1 Computational Genomics Course Lecture 12 fall 2002/03 School of Computer Science Tel-Aviv University Instructor: Benny Chor Many slides taken (with permission…)

7

Physical MapsThe ultimate physical map is the complete sequence of a chromosome.

6

Page 8: 1 Computational Genomics Course Lecture 12 fall 2002/03 School of Computer Science Tel-Aviv University Instructor: Benny Chor Many slides taken (with permission…)

8

A somatic cell technique that is used for ordering markers along a chromosome and estimating the physical distances between them.

Markers are genomic sequences of length approx. 200bp,appearing uniquely on the human genome.

Markers’ locations and relative order is unknown a-priory, but should become known after the RH experiments.

Analyzing the experimental data is a challenging and demanding computational task.

Radiation Hybrid (RH) Mapping

7

Page 9: 1 Computational Genomics Course Lecture 12 fall 2002/03 School of Computer Science Tel-Aviv University Instructor: Benny Chor Many slides taken (with permission…)

9http://www1.genetics.wisc.edu/466/fall99/lect13/sld011.htm

Donor cells (haploid or diploid) are irradiated, and chromosomes breaks at random locations.

The irradiated cells are fused with non-radiated rodent cells.

Hybrids are formed.

Radiation Hybrid (RH) Mapping

8

Page 10: 1 Computational Genomics Course Lecture 12 fall 2002/03 School of Computer Science Tel-Aviv University Instructor: Benny Chor Many slides taken (with permission…)

10

Marker Retention Patterns

Experiment output: An n-by-m matrix, indicating which marker is retained in which hybrid cell.

Approx. n ~ 100 “hybrids” are used for mapping m ~ 150 markers on one chromosome.

The resulting cells are screened for the presence or absence of the markers.

9

Page 11: 1 Computational Genomics Course Lecture 12 fall 2002/03 School of Computer Science Tel-Aviv University Instructor: Benny Chor Many slides taken (with permission…)

11

Radiation Hybrid (RH) Mapping Input

Example data for radiation hybrid mapping:

Markers:+ (1) presence - (0) absence

10

Page 12: 1 Computational Genomics Course Lecture 12 fall 2002/03 School of Computer Science Tel-Aviv University Instructor: Benny Chor Many slides taken (with permission…)

12

RH Pannels & Hamming Distances

V = 10100110001U = 10010001000W = 01101110111

U, V, W are three RH panels (binary vectors) of length 11.The Hamming distance between V and U is 6.The Hamming distance between U and W is 11.

11

Page 13: 1 Computational Genomics Course Lecture 12 fall 2002/03 School of Computer Science Tel-Aviv University Instructor: Benny Chor Many slides taken (with permission…)

13

RH Computational Task

Intuition: Close-by markers will be retained or lost together. Far away markers retained or lost independently. The further apart two markers are, the more likely it is that radiation will break between them, resulting in two separate chromosomal fragments. Viewing each marker as length m binary vector, Hamming distance between vectors is indicative of markers’ distance on chromosome.

Input: n-by-m 0/1 matrix (one row per marker).Desired Output: Ordered markers (a permutation on {1,2,…,n})

12

Page 14: 1 Computational Genomics Course Lecture 12 fall 2002/03 School of Computer Science Tel-Aviv University Instructor: Benny Chor Many slides taken (with permission…)

14

RHO - Radiation Hybrid OrderingAmir Ben-Dor, Benny Chor and Dan Pelleg, Dept. of Computer Science, Technion.

http://www.cs.technion.ac.il/~benny/RHO.pdf

A software package that implements a number of heuristicsthat attempt to order genomic markers along the chromosome,given as input the results of an RH biological experiment.

The heuristics are based on formulating an appropriate optimization problem, “reducing” RH to the traveling salesmanproblem (TSP).

Two different optimization problems:1. Nonparametric: Minimum obligate chromosome break (MOB), a problem of “combinatorial nature”. 2. Parametric: Maximum likelihood estimation (MLE), a problem of “statistical nature”. 13

Page 15: 1 Computational Genomics Course Lecture 12 fall 2002/03 School of Computer Science Tel-Aviv University Instructor: Benny Chor Many slides taken (with permission…)

15

RHO - Radiation Hybrid OrderingAmir Ben-Dor, Benny Chor and Dan Pelleg, Dept. of Computer Science, Technion.

http://www.cs.technion.ac.il/~benny/RHO.pdf

TSP is obviously NPH, but its symmetric version is amenable tovery efficient hueristics.

Two hueristic approaches:

1. Simulated annealing: Finds a tour (upper bound), typically fairly close to optimal (but how does one know that…?) 2. Held-Karp: Finds a minimum spanning tree (lower bound). By modifying the underlying graph, MST becomes more and more like a path.3. Comparing results of (1) and (2) we get a good estimate of accuracy of solution. If both are same, this proves optimality of solution ! 14

Page 16: 1 Computational Genomics Course Lecture 12 fall 2002/03 School of Computer Science Tel-Aviv University Instructor: Benny Chor Many slides taken (with permission…)

16

RHO Output

15

Page 17: 1 Computational Genomics Course Lecture 12 fall 2002/03 School of Computer Science Tel-Aviv University Instructor: Benny Chor Many slides taken (with permission…)

17

RHO ResultsChromosome 1 analysis: There are 132 markers and 93 hybrids.Results for the Maximum Likelihood estimation method:

WhiteheadInstitute datavsRHO

16

Page 18: 1 Computational Genomics Course Lecture 12 fall 2002/03 School of Computer Science Tel-Aviv University Instructor: Benny Chor Many slides taken (with permission…)

18

Other Example:

17

In this example, RHO ordered each of the two “arms” of the chromosome in accordance with the original order. However, the first “arm” was reversed. This phenomena is fairly common,probably indicating that retention on one side of the centromeris independent of retention on the other side.

Page 19: 1 Computational Genomics Course Lecture 12 fall 2002/03 School of Computer Science Tel-Aviv University Instructor: Benny Chor Many slides taken (with permission…)

19

Comparison of Genetic Map and RH Map of Chromosome 8q24

Geneticmap

RHmap

Common sites

Adapted from: Lewis et al., 1995 18

Page 20: 1 Computational Genomics Course Lecture 12 fall 2002/03 School of Computer Science Tel-Aviv University Instructor: Benny Chor Many slides taken (with permission…)

20

http://corba.ebi.ac.uk:80/RHdb/index.html

Release 19.0 - 22 January 2001. This release contains: 133239 RH entries (106574 different STSs), 92 maps, 15 Panels, 229 experimental conditions and 3 species.

Radiation Hybrid Database

http://www.ncbi.nlm.nih.gov/dbEST/ 19

Page 21: 1 Computational Genomics Course Lecture 12 fall 2002/03 School of Computer Science Tel-Aviv University Instructor: Benny Chor Many slides taken (with permission…)

21

Map Viewer: A tool for visualizing whole genomes or single chromosomes.

• Where does a particular gene exist within an organism's genome ? • Which genes are located on a particular chromosome and in what order ?

• Other links from Entrez, for a gene that exists in a particular chromosomal region ?

• What is the distance between genes ?

NCBI Map Viewer - A Tool for Integrating Genetic and Physical Maps

http://www.ncbi.nlm.nih.gov/cgi-bin/Entrez/map_search?chr=hum_chr.inf&query

20

Page 22: 1 Computational Genomics Course Lecture 12 fall 2002/03 School of Computer Science Tel-Aviv University Instructor: Benny Chor Many slides taken (with permission…)

22

NCBI Map View

Scaleon chromosome

http://www.ncbi.nlm.nih.gov/cgi-bin/Entrez/hum_srch?chr=hum_chr.inf&query

21

Page 23: 1 Computational Genomics Course Lecture 12 fall 2002/03 School of Computer Science Tel-Aviv University Instructor: Benny Chor Many slides taken (with permission…)

23

NCBI Map View

Scaleon chromosome

See also Map View at:http://genome.ucsc.edu/

22