Transcript Chapter 1 PowerPoint Slides

Discovering Genomics, Proteomics, & Bioinformatics
second edition
by A. Malcolm Campbell and Laurie J. Heyer
Chapter 1: What’s Wrong with My Child?
1
2
• Begins as hypothetical dialogue as if you
are a clinician seeing a patient
• One of your patients is a 6yo boy (BB)
–
–
–
–
Chapter 1
digestive problems
child is unresponsive to questions
stomach ache for several months
unable to stand or walk on his own
3
histology -- the microscopic study of thinly sliced tissues/cells
Figure 1.1
4
Genotyping
• Collect blood and genotype family (RFLPs)
• Another patient and family (girl GG)
• GG's phenotype -- some similar symptoms
–
–
–
–
–
Cannot walk
enlarged calves
poor reflexes in limbs
no intellectual deficits
3 healthy brothers
5
Figure 1.2
markers != genes
markers may be very near genes
(even within introns)
Easy to mistakenly equate genes
6
with markers -- I do it.
The disease -- recessive?
• BB's disease is recessive (on X chrm)
• Perform karyotype
• Why is GG affected -- she has TWO copies
of X chromosome?
• Markers suggest that she only has one copy
of the disease haplotype (or bad copy of
defective gene near markers)
7
Figure 1.3a - c
8
Figure 1.3d - e
9
Mosaics
• GG
– wild-type X chrm appears to be replicating later
than mutant (approx 55 out of 56 times)
– X inactivation
– 1 time out of 56, the wild-type X chromosome
is expressing genes
– Why?
10
Figure 1.4
11
Sequence Them
• Let's assume we know how to sequence
them (more on that later).
12
How can GG have an "R" if neither mother or father has it?
Table 1.1
13
•
•
•
•
•
•
Duchenne's muscular dystrophy
(DMD)
mRNA is > 14,000
gene is ~ 1 million nts
How is this possible? 14,000 != 1,000,000
Back in the day, 500 bp by hand, per day
Impossible
Gene later became known as "dystrophin"
– Gene names, and aliases are still a big problem
– May actually be more misleading
– A novel gene may be named because it "looks" like
another gene.
– may have alternative splicing
– The function thought to be associated with original gene may be wrong,
etc.
• "chaperonin-like"
14
40 years of research…in < 20 slides
• Presents as nice linear and logical sequence of events.
Papers and case studies are often presented like this, but
the reality is that there were many failed experiments and
hypotheses. People came and went -- real science is
almost never this linear
• Writing grants
• Teaching and faculty meetings
• Meetings and conferences
• Training students, staff, scientists, secretaries, etc.
• Writing papers
• Experiments, experiments, experiments
15
Math Minute 1.1
•
•
•
•
•
•
•
E-values and BLAST (we'll come back to BLAST later)
BLAST -- basic local alignment search tool
BLAST (in this case BLASTp for "protein") takes as input one protein
sequence (query) and compares it to a database of multiple proteins sequences
(subjects)
A significant hit between query and subject is measured by E-value (expect
value)
– the E-value is the value you'd expect from a hit in two sequences by
chance
– two extremes
• if your database is so large you have all possible combinations of
sequences -- then you are VERY likely to find the sequence you are
looking for -- but this E-value will be very low (I.e. good or
significant) since there are so many sequences
• if your database is very small, then a "hit" will not be as significant
(and rarely will you find a match).
– As E-value gets closer to 0 -- the hit becomes more significant
A "bit score (S)" is a measure of the similarity between Q and S.
Big bit score corresponds to small e-value
16
Essence of BLAST (and BLAT) is to "hash"
What's next?
•
•
Need an animal model.
Why?
• Mutant mice
mdx
– develops DMD
– ID'ed by biologists and isolated
– mouse labs to this day watch for and cultivate spontaneous phenotypes
(Jax lab)
– had mouse before the gene -- no guarantee it’s the same gene
– may actually try therapies
– gene found (with anti-body)
•
•
•
•
dystrophin localized to surface of cell in skeletal muscles
in neurons (explains mental retardation)
no dystrophin in mdx mice
explains why its recessive -- loss of function (remember mom had one good
copy, and one bad copy
17
Utrophin
• Looks very similar to dystrophin (based on
sequence similarity)
• expressed in fetal tissues, regenerating
muscle, and other tissues during
development (ubiquitous)
• small-caliber muscles (like eye-movement)
18
For more information, wikipedia "polyclonal antibody" or
"monoclonal antibody"
19
Immunofluorescent labeling of dystrophing and utrophin localizes
protein to cell membrane (and neurons -- data not shown).
Figure 1.5
20
Function of dystrophin?
• Jim Ervasti and Kevin Campbell
• Identified proteins/complexes associated
with dystrophin
21
dystrophin -- predicted to form "coiled-coil"
motif -- ?
X-ray diffraction crystallography
Figure 1.6
22
Now thought that this structure
links cytoskeleton (cells) to
skeleton (bones, muscles,
tendons too)
Figure 1.7
23
Systems biology
Figure 1.8
reductionism to synthesis
24
DMD -- Muscle Deterioration?
• How does mutation cause disease?
• How does single missense cause children to
develop disease?
• Why children and not infants?
• Why do the muscles gradually deteriorate?
25
DMD
54 L --> R (at amino acid position 54, leucine
has been changed to arginine)
Is this a big change?
Hypothesis?
26
Hypothesis
• Hypothesis: muscle contracts and damages
the cell membrane
– because cell is anchored to the bone, but not to
the cytoskeleton (rigid structure of the cell).
• Infants don't move around as much -- so
damage doesn't occur
• Explains muscle deterioration, but what
about neurons?
27
DMD -- neuronal?
• penetrance -- mental symptoms of DMD
has low penetrance (not everyone with the
mutation gets the symptom).
• suggested that maybe neuronal cells fail to
attach to each other, or other interfaces
• not really sure…?
28
• Stopped here
S
29
More patients
• mother
• son (YY) -- 4 yo
• daughter (DD) -- 12 yo
30
Unlikely that
both children
would inherit
4 disease alleles.
Why?
Figure 1.9
31
Mutated betasarcoglycan.
Misforming of
complex
Figure 1.10
32
Only 3 mutations found in sarcospan -- in 50 families -- and it
did not cause the disease.
Figure 1.11
33
Truncated sarcoglycan
Figure 1.12
34
Table 1.2
35
BBS
• Chiang AP, Beck JS, Yen HJ, Tayeh MK, Scheetz TE,
Swiderski RE, Nishimura DY, Braun TA, Kim KY, Huang
J, Elbedour K, Carmi R, Slusarski DC, Casavant TL, Stone
EM, Sheffield VC.
Homozygosity mapping with SNP arrays identifies
TRIM32, an E3 ubiquitin ligase, as a Bardet-Biedl
syndrome gene (BBS11).
Proc Natl Acad Sci U S A. 2006 Apr 18;103(16):6287-92.
Epub 2006 Apr 10.
PMID: 16606853 [PubMed - indexed for MEDLINE]
36
Search for all mutations in calpain.
Figure 1.18a
37
Figure 1.18b-d
38
End chapter 1
• Although the rest of the material is
extremely interesting, and I encourage you
to read if interested, it gets somewhat
speculative. So you will not be responsible
for this material.
39
Figure 1.13
40
Figure 1.13a
41
Figure 1.13b
42
Figure 1.13c
43
Figure 1.14
44
Figure 1.15
45
Figure 1.16
46
Figure 1.17
47
Figure 1.18
48
Figure 1.18a
49
Figure 1.18b-d
50
Figure 1.19
51
Figure 1.20
52
Figure 1.21
53
Figure 1.22
54