The Pines - Davidson College

Download Report

Transcript The Pines - Davidson College

Genomic Medicine
A. Malcolm Campbell
James G. Martin Genomics Program Director
Professor of Biology, Davidson College
The Pines
October 28, 2013
Outline of Talk
1. What is a genome?
2. How to sequence genomes?
3. Diagnose and Treat Cancers Better?
4. New Drugs from Failures: Iressa
5. New Treatment Paradigms
6. Gut microbiome
Science Presentation
Give you the data, help you interpret.
What is a Genome?
What is a Genome?
• Genetic information that you unique
• Human genome 3.4 billion base pairs
• G:C or
A:T base pairs
• ~23,000 genes
What is the Human Genome?
If the human genome were compiled in books:
• 200 volumes, 1000 pages each
• read 10 bases/second = 315,360,000 bases/year
• 9.5 years to read out loud (without stopping)
How do you sequence a genome?
• Determine order of bases on all 23 (24) chromosomes
• Can only read 30 to 700 bases at a time
• Cannot sequence a genome in one run
• “Whole Genome Shotgun” sequencing
How do you sequence a genome?
• Determine order of bases on all 23 (24) chromosomes
• Can only read 30 to 700 bases at a time
• Cannot sequence a genome in one run
• “Whole Genome Shotgun” sequencing
Modern Genome Sequencing
Modern Genome Sequencing
This is an
analogy for
genome
assembly. This
page will be
torn into
horizontal
strips.
Modern Genome Sequencing
is an
s an
This i
Modern Genome Sequencing
This i
is a
s an
Modern Genome Sequencing
This i
is a
s an
This is an
analogy for
genome
assembly. This
page will be
torn into
horizontal
strips.
You don’t know the language or syntax.
This i
is a
s an
Huge Pile of Strips
This i
is a
s an
Needle in a Haystack?
This i
is a
s an
Reassemble the Tree from Paper
Modern Genome Sequence Assembly
multiple copies of genome
aligned sequencing reads
consensus genome sequence
Modern Genome Sequence Assembly
multiple copies of genome
aligned sequencing reads
consensus genome sequence
Modern Genome Sequence Assembly
multiple copies of genome
aligned sequencing reads
high coverage
consensus genome sequence
low coverage
Modern Genome Sequence Assembly
multiple copies of genome
aligned sequencing reads
consensus genome sequence
Reference Genome Assembly
ATGGCATTGCAA
TGGCATTGCAATTTG
AGATGGTATTG
GATGGCATTGCAA
GCATTGCAATTTGAC
ATGGCATTGCAATTT
AGATGGTATTGCAATTTG
Reference Genome Assembly
ATGGCATTGCAA
TGGCATTGCAATTTG
AGATGGTATTG
GATGGCATTGCAA
GCATTGCAATTTGAC
ATGGCATTGCAATTT
AGATGGTATTGCAATTTG
consensus
AGATGGCATTGCAATTTGAC
How do I remember them all?
206 bones
> 60 organs
Think like a genomicist…
206 bones
> 60 organs
1 – 22 + X & Y
Even Easier
Is there a better way to diagnose
and treat cancers?
Is there a better way to diagnose
and treat cancers?
Diffuse Large B Cell Lymphoma
(DLBCL)
Diffuse Large B Cell Lymphoma
(DLBCL)
25,000 new cases each year in America
Half of patients die despite chemotherapy
Why subject them to chemo if not helpful?
Who will benefit from chemotherapy?
Brown and Botstein
Who will benefit from chemotherapy?
patient and control biopsies
control set
Measure Gene Activity in All Cells
Characterize Cancers by Gene Activity
Retrospective Clinical Outcomes
based on gene activity
Retrospective Clinical Outcomes
based on gene activity
traditional prediction
Retrospective Clinical Outcomes
wrong 27%
wrong 32%
Retrospective Clinical Outcomes
wrong 37%
Resort Low Risk Patients
wrong 37%
wrong 29%
Six Key Indicator Genes
Genes On
Genes On
LMO2
BCL6
FN1
BCL2
SCYA3
CCND2
chemo
no chemo
Can Breast Cancer Treatment Be Improved?
Can Breast Cancer Treatment Be Improved?
tissue biopsies
Can Breast Cancer Treatment Be Improved?
No Patterns Emerged
control set
patient biopsies
People Are More Unique Than Their Cancers
before (BE) and after (AF) chemotherapy
Characterize Cancers by Personal Differences
Characterize Cancers by Personal Differences
“Breast Cancer” is a Misnomer
There are at least 5 Different Breast Cancers
No such thing as THE cure for cancer.
The Discovery of Iressa
non-small cell lung cancer
Success from Failure: Iressa
Cancer Treatment from Nature’s Oddities
Naked Mole Rat (NMR)
Face That Only A Mother Could Love
Face That Only A Mother Could Love?
Cancer Treatment from Nature’s Oddities
lives 40 years
never has cancer
lives 4 years
tumors common
What prevents cancer in naked mole rats?
high-molecular-mass hyaluronan (HA),
What prevents cancer in naked mole rats?
5 times bigger than human HA
lower degradation rate than human
HA Over Produced in NMR Tissues
HA Over Produced in NMR Tissues
HA Over Produced in NMR Tissues
relative amount of HA
HA Over Produced in NMR Tissues
Block HA  Tumors Form
mouse cells
neg. control
carcinogen 1
carcinogen 2
carcinogen 3
NMR
NMR - HA
Block HA  Tumors Form
mouse cells
neg. control
carcinogen 1
carcinogen 2
carcinogen 3
NMR
NMR - HA
Block HA  Tumors Form
mouse cells
neg. control
carcinogen 1
carcinogen 2
carcinogen 3
NMR cells
NMR - HA
Block HA  Tumors Form
mouse cells
neg. control
carcinogen 1
carcinogen 2
carcinogen 3
NMR cells
NMR no HA
Inject Mice with NMR Cells +/- HA
injection
mouse tumor
NMR cells
NMR cells
++ HAase
NMR cells
HA not made
recipient
tumors
tumors
Inject Mice with NMR Cells +/- HA
injection
mouse tumor
NMR cells
NMR cells
++ HAase
NMR cells
HA not made
recipient
tumors
Inject Mice with NMR Cells +/- HA
injection
mouse tumor
NMR cells
NMR cells
++ HAase
NMR cells
HA not made
recipient
tumors
Inject Mice with NMR Cells +/- HA
injection
mouse tumor
NMR cells
NMR cells
++ HAase
NMR cells
HA not made
recipient
tumors
Inject Mice with NMR Cells +/- HA
injection
mouse tumor
NMR cells
NMR cells
++ HAase
NMR cells
HA not made
recipient
tumors
Inject Mice with NMR Cells +/- HA
injection
mouse tumor
NMR cells
NMR cells
++ HAase
NMR cells
HA not made
recipient
tumors
Inject Mice with NMR Cells +/- HA
injection
mouse tumor
NMR cells
NMR cells
++ HAase
NMR cells
HA not made
recipient
tumors
Inject Mice with NMR Cells +/- HA
injection
mouse tumor
NMR cells
NMR cells
++ HAase
NMR cells
HA not made
recipient
tumors
Cancer Treatment from Nature’s Oddities
Remember this next time Congress makes fun
of scientists studying odd creatures.
Human Body Is Mostly Non-Human
50 trillion human cells
Human Body Is Mostly Non-Human
50 trillion human cells
500 trillion non-human cells
Human Body Is Mostly Non-Human
50 trillion human cells
500 trillion non-human cells
10% human, 90% bacterial
Define the Human Microbiome
27 sites
9 adults
4 occasions
Define the Human Microbiome
each sample
colored by location
clustered by similar species
Human Microbiome by Location
Human Microbiome by Location
Human Microbiome by Location
Human Microbiome by Location
Human Microbiome by Location
Two Types of Skin Habitats
Human Microbiome by Location
Human Microbiome Variations
Human Microbiome Variations
Microbiome Transplant Experiments
compare to natives
compare to transplants
Human Microbiome Variations
compare to natives
compare to transplants
Human Microbiome Variations
forearm sustains transplants
forehead replaces transplants
Human Microbiome
many skin sites more diversity than gut
high-diversity skin sites
• forearm
• palm
• index finger
• back knee
• sole of the foot
Myth Busters: dog’s mouth is really clean
Myth Busters: dog’s mouth is really clean
Medical Uses of Microbiome
mice fed low doses of antibiotics long term
15% more body fat than the control mice,
Medical Uses of Microbiome
sequenced bacteria in stool samples humans
• 55 were thin
• 122 overweight or obese
Medical Uses of Microbiome
obese people lack bacterial diversity in guts
antibiotics reduce microbiome diversity
Medical Uses of Microbiome
obese people lack bacterial diversity in guts
missing microbes methane producers,
“the carbon that does not get out
as gas could be incorporated as fat.”
Medical Uses of Microbiome
Using six species, predict obese 80% of the time
Predictions based on genetics only 58% of the time
Guess what is being tested now…
Guess what is being tested now…
fecal transplants
Thanks to my Colleagues
Laurie Heyer
Leland Taylor ‘12
supported by
James G. Martin Genomics Program
Davidson College