High resolution mapping of mammalian regulatory loci using Radiation Hybrids:

1
High resolution mapping of mammalian regulatory
loci using Radiation Hybrids

• Pan, Tuo
• Basic Medical Sciences
• Zhejiang University
• September 14th, 2006
• Under the guidance of
• Chris Park
• In the lab of
• Desmond Smith
• Molecular Medical Pharmacology, UCLA

2
Goal of the Project

• To map loci regulating the expression of any
  given gene in the human genome
• At high resolution
• Map networks of gene regulation

3
Methods

• How are we going to accomplish this?

4
Radiation Hybrid Cell Lines
10,000 radsof X rays
Fragmented DNA ofIrradiated DonorHuman Cell
Diploid Male LymphoblastoidDonor Human Cell
Polyethylene Glycol
RecipientHamster Cell
Radiation Hybrid Cell Line 17 Human DNA 100
Hamster DNA
5
Methods

• How are we going to accomplish this?

6
Microarray Analysis

• Transcriptional profiling data collected for all
  83 radiation hybrid cell lines
• Map loci responsible for transcription of all
  genes

7
Methods

• How are we going to accomplish this?

8
Linkage Analysis
Example pairwise comparison of Gene no. 1 and
Marker no. 1
These RH Cells (83) show Marker no. 1 is present
and have high Gene expression.
Expression Level of Gene no.1
These RH Cells (17) showMarker no. 1 is absent
andhave low Gene expression.
0
1
Cells with or without Marker no. 1
9
Linkage Analysis
Example pairwise comparison of Gene no. 1 and all
120 Markers
Expression
Expression
Expression
0
1
0
1
0
1
Marker no. 1
Marker no. 2
Marker no. 3
Hotspots (high LOD)
LOD
0
72
96
48
24
120
All 120 Markers placed in order on genome
10
Methods

• How are we going to accomplish this?

11
System Radiation Hybrid Panel

• 83 Cell Lines
• Each with a different set of donor fragments
• Each line contains
• 100 of Hamster Genome
• 17 of Human Genome in random fragments
• Enriched in the Thymidine kinase (Tk) locus

1
2
3
4
5
7
6
8
15
16
19
17
18
20
21
22
XY
12
Radiation Hybrid Panel

• Originally used for physical mapping of genomes
• Can assemble a map with a marker order that is
  most likely based on PCR screening data
• Marker a landmark (DNA sequence) that defines a
  unique location on a chromosome
• Distance between markers can be estimated
• Fine breakages enable extremely precise
  localization of markers on the genomic map

13
RH Cell Lines have been
densely genotyped

• Are these cell lines stable?

14
Question time!

• Q When were those RH genotype data collected?
• A HmmMore than 10 years ago.
• Q 10 years ago? Thats a long time! Havent the
  cell lines changed so far?
• A Well, I guess so. You know genetic stuffs are
  usually very stable
• Q No guess in science. Show me the evidence,
  please.
• A Okay, look at this

15
Our PCR genotyping process
16
Stanford RH-G3 Panel
83 Cell Lines
Each cell fused with a different set of
fragmented human DNA
120 Markers
Publicly available data showing absenceor
presence of each marker in each cell line
Location of each marker on human genomeis
determined
17
Results
18
Genotyping Data (part)
19
Discussion
93.82
Total Concordance Ratio

• Shows that after multiple generations, RH cell
  lines are still stable and have a high
  concordance with the public data they got more
  than one decade ago.

20
Roadmaps

• 1st-4th week (July 10th-August 3rd)
• Accomplished PCR screening
• 120 primer (public) samples
• 83 radiation hybrid cell lines together with 3
  controls
• 5th week (August 3rd-August 9th)
• Programming fundamental data analysis programs
• Detecting out strange results
• Switched cell lines 62 and 66
• 71 probably have genetic changes
• several cell lines have interesting results
  versus public data
• 6th-8th week (August 10th-August 31st)
• Re-doing some experiments based on our previous
  results
• Found cell line 71 has been contaminated
• 9th-10th week (September 1st-September 14th)
• Designing and writing for ZJU-UCLA 2006 Website
• Some other paperwork

21
Suggestions

• Improvement on cell line 71.
• Can expect an enhancement of total concordance
  ratio from 91.98 to 92.37
• Switch the cell line of 62 and 66.
• Can expect another enhancement of total
  concordance ratio from 92.37 to 93.82
• Have a check over those U-bands (unknown
  results).
• Now we still have almost 200 U-bands, with part
  of them rechecked.
• Among those rechecked ones, almost all of them
  still kept the original results.
• Use more data analysis tools to find out a better
  way figuring out our dates.
• Use more graphic ones so as to get a more
  intuitionistic understanding.

22
Future Analyses

• Replicate experiments with a different radiation
  hybrid panel
• Pooling data results in higher resolution mapping
• Biologically validate our experimental data
• Transfect candidate regulatory genes and monitor
  expression of target genes.

23
References

• Elizabeth A. Stewart, Kathleen B. McKusick, Amita
  Aggarwal, Ewa Bajorek et al. 1997. An STS-Based
  Radiation Hybrid Map of the Human Genome. Genome
  Research 7 (5) 422
• Linda C. McCarthy, Jonathan Terrett, Maria E.
  Davis et al. 1997. A First-Generation Whole
  Genome-Radiation Hybrid Map Spanning the Mouse
  Genome. Genome Research 7 (12) 1153
• John E. J. Rasko, Jean-Luc Battini, Leonid
  Kruglyak et al. 2000. Precise gene localization
  by phenotypic assay of radiation hybrid cells.
  Proceedings of the National Academy of Sciences
  97 (13) 7388
• Susan R. Ross, Jason J. Schofield et al. 2002.
  Mouse transferrin receptor 1 is the cell entry
  receptor for mouse mammary tumor virus.
  Proceedings of the National Academy of Sciences
  99 (19) 12386

24
Thanks!
Special thanks to

• China
• Zhejiang Univ.
• ZJU-UCLA 2006
• Dr. Sun
• Dr. Wang
• Teacher Cao
• All my fellow friends

• Smith Lab
• Chris Park
• Arshad Khan
• Albert Chen
• Mark Chin
• Richard Wang
• Alex Geng
• George
• Desmond Smith

25
Big Happy Family
26
(No Transcript)
27
Microarray Analysis
20,000 to 25,000 Genes
Cell Line 1
Cell Line 2
- The colors identify levels of mRNA abundance
or gene expression. - Some genes are expressed
more than others.
Cell Line 3
Cell Line 83