Gene expressions analysis by massively parallel signature
Download
Report
Transcript Gene expressions analysis by massively parallel signature
Gene expressions analysis by massively
parallel signature sequencing (MPSS) on
microbead arrays
Sydney Brenner et al.
Jan 26, 2007
DNA-basic definitions
Nucleotide (Base) - consists of a sugar, phosphate
and a base
Base-Pair Rule – Hydrogen bonds (A-T, C-G)
Gene - a segment of DNA that codes for a protein,
which in turn codes for a trait (skin tone, eye
color,…etc), a gene is a stretch of DNA.
DNA Polymerase - Enzymes that catalyze the
polymerization of deoxyribonucleotides alongside a DNA
strand, which they "read" and use as a template. The
newly-polymerized molecule is complementary to the
template strand and identical to the template's partner
strand.
DNA Sequencing
Sequencing: determining the nucleotide order of a
given DNA fragment, called the DNA sequence.
Sanger Method (’75): Chain termination
•
Prepare single strand DNA (heating)
•
Add a mixture of deoxy nucleotides (dATP,
dGTP, dCTP, dTTP)
•
Add a mixture of dideoxy nucleotides
(ddATP,ddGTP,ddCTP, ddTTP)
•
Add DNA polymerase I
A lot more deoxynucleotides than
dideoxynucleotides
•
Sort based on length (gel)
Animation:
http://www.dnalc.org/ddnalc/resources/sangerseq.html
Restriction Enzymes
• 1978 Nobel Prize in medicine (awarded to Werner
Arber, Daniel Nathans, and Hamilton Smith)
• Enzymes that cut double stranded DNA
• The cleaved chemical bonds can be reformed by
ligases
• Restriction enzyme cuts only double-helical segments
that contain a particular nucleotide sequence (i.e.
recognition sequence)
• Types of Restriction enzymes: I, II, III:
I,III: recognize specific sequences but the cleavage
sites are at variable distances
II: cleavage occurs at specific sites at or close to the
recognition sequence
Fluorescence Activated Cell Sorting (FACS)
First cell sorter: Mack Fulwyler (1965)
Expanded by Len Herzenberg
Cells are tagged by antibodies linked to
fluorescent dye. The antibody is bound to a
protein that is uniquely expressed in the
cells that we want sorted.
The nozzle vibrates to form drops which
contain single cells
Electrical charge is used to sort the cells
Motivation
Goals (upon determination of human genome):
1.
To discover and understand the function and variation of genes
2.
How do these qualities affect health and disease?
Available techniques:
1.
Hybridization of probes into microarrays
Advantages: large scale, capable of detecting a wide range of gene
expression levels.
Disadvantages: variability due to probe hybridization, cross reactivity,
element to element differences, and microarray to microarray
differences
Counting of tags or signatures of DNA fragments
2.
Advantages: Statistically more robust, don’t require standardization or
repetition, precision and accuracy can be increased by increasing the
size of the sample
Disadvantages: Difficult to realize routinely and not cost effective
Massively Parallel Signature Sequencing (MPSS)
Cloning on microbeads
1% of tags
# of transcripts:
3-4e4
# of oligonucleotides: 1.67e7
# of conjugates: 5-7e11
We want to make sure that # of tags is at least a 100 times the number of templates
this will ensure that if we take 1% of tags, we have a sample where all DNA’s are
represented and they all have unique tags with at least 99% probability.
Massively Parallel Signature Sequencing (MPSS)
PCR is used to amplify the sample. The resulting single stranded
DNA’s are hybridized with a population of microbeads. (note: 1% of
microbeads are loaded)
Separate loaded microbeads from unloaded ones using FACS.
Each microbead has a population of 104-105 identical copies of a
single kind of template molecule
Massively Parallel Signature Sequencing (MPSS)
Decoding…
1
TNNN…Fx1
NTNN…Fx2
NNCN…Fx3
TTCC
NNNC…Fx4
Ligate 1024 encoders
43 x 16
Decoders for Fx1…4
2
3
16x decoding per cycle….repeat
Massively Parallel Signature Sequencing (MPSS)
Massively Parallel Signature Sequencing (MPSS)
Comparison to conventional methods
Compared to PE Biosystems, model
377 DNA Sequencer
Discussion
Method does not require separation of fragments to
generate sequence information
Time series of spatially localized microbeads (can pack
beads closely in monolayers)
The main advantage: parallel nature of the processmillions of templates can be handled together without
need for separation. (Ideal for gene expression)
Conventional sequencing: analyze thousands of
templates to give sequences with 100s of bases
MPSS: analyze millions of templates to give sequences
of length few 10s of bases
Critique Summary
Strong points:
A powerful and innovative technique
Lots of work done to prove the functionality
Weak points
Not the clearest paper (too much information compacted in a few
sentences)
There are more efficient ways of encoding/decoding the adaptors
The figures are better shown in a different order
It would be good to include some more information about the
cloning of the microbeads
I could use more comments on how good the results are (is it just
me!?)
Better encoding (Hansen’s method)
0
1
2
3
4 5 6 7
0
0
0
0
0
0
0
0
0 0 0 0
0 0 0 1
AAAC 0
AAAG 0
0
0
0
0
0
0
0 0 1 0
0 0 1 1
AAAA
AAAT
.
.
.
AAAT------------7
AAAG------------6,7
23 = 8 different encoders required as opposed to 16 used here.