Transcript Powerpoint slides

Evaluation of Affymetrix array
normalization procedures
based on spiked cRNAs
Andrew Hill
Expression Profiling Informatics
Genetics Institute/Wyeth-Ayerst Research
Outline
• The GI/Harvard C. elegans array dataset as a
normalization testbed
• Some general challenges of array data reduction
• GeneChip Scaled Average Difference (ADs)
– the constant mean assumption
• A purely spike-based normalization strategy
(Frequency)
• A hybrid normalization (Scaled Frequency)
• Conclusions
October 11, 2001
2
GI/Harvard C. elegans dataset
•
•
•
This data set used to evaluate several normalization procedures
Experiments:
– 8 developmental stages of the worm C. elegans were profiled,
ranging from egg to adult worm
– n=2-4 replicate hybridizations for most array designs at most
stages
– 52 total arrays
Arrays:
– Three custom worm GeneChip designs (A, B, and C)
– Each array monitors between 5700-6700 ORFs, in aggregate ~98%
of the worm genome
– Chip A: ORFs with cDNA/EST matches in AceDB
– Chips B/C: other ORFs
– Several worm ORFs tiled on all 3 arrays for across-array-design
comparisons
Science 290 809-812; Genome Biology (in the press)
October 11, 2001
3
Some challenges of Affymetrix
GeneChip data reduction
• Array data from Affymetrix GeneChip sofware (pre-MAS 5.0):
– negative low intensity signals
– lack of across-design normalization standard
– limited QC information
• Spike-based normalization methods can help to address each of
these challenges
Normalization: array scaling of average difference data from multiple
arrays/designs to minimize technical noise among arrays
• Current “standard” normalization procedure is a global scaling
procedure: the GeneChip scaled average difference (ADs)
October 11, 2001
4
GeneChip Scaled Average
Difference (ADs)
• The trimmed (2%) mean intensity of all
probesets on all arrays is scaled to a
constant target level.
• Works well in many cases (e.g.
replicates)
• Some obvious situations where the
“constant mean assumption” may not be
well supported.
October 11, 2001
5
Constant mean assumption:
problematic cases
•Chips monitoring a
“small” fraction of
transcriptome
•Non-random gene
selection on arrays (e.g.
C. elegans A vs. B/C)
•Large biological
variation in expression
October 11, 2001
6
A cRNA spike-based normalization
procedure (Frequency)
• Add 11 biotin-labeled cRNA spikes to
each hybridization cocktail
• Construct a calibration curve
• Use the Absent/Present calls for the
spikes to estimate array sensitivity
• Dampen AD signals below the
sensitivity level to eliminate negative AD
values.
October 11, 2001
7
Eleven spiked cRNAs
Spiked Transcript ATCC Accession Affymetrix Gene Qualifier Final concentration (pMol) Final concentration (ppm)
DAPM
87826
AFFX-DapX-M_at
30
950
DAP5
87827
AFFX-DapX-5_at
10
317
CRE5
87832
AFFX-CreX-5_at
5
158
BIOB5
87825
AFFX-BioB-5_at
2.5
79
BIOD3
87830
AFFX-BioDn-3_at
1.2
38
BIOB3
87828
AFFX-BioB-3_at
0.6
19
CRE3
87835
AFFX-CreX-3_at
0.4
13
BIOC5
87833
AFFX-BioC-5_at
0.3
10
BIOC3
87834
AFFX-BioC-3_at
0.2
6
DAP3
87831
AFFX-DapX-3_at
0.15
5
BIOBM
87829
AFFX-BioB-M_at
0.1
3
October 11, 2001
8
Response to spikes over 2.5 log range
Figure 2
•Fit response with
S-plus GLM,
gamma error
model, zero
intercept.
•Power law fit
AD=kFn yields
n=0.93
•cRNA mass,
scanner PMT gain
are important
determinants of
response
October 11, 2001
9
1.0
Chip sensitivity calculation
P P
P
P
P
P
P
0.6
0.4
0.2
0.0
A/P call
0.8
P
A A
0
1
A
2
3
4
log(frequency)
October 11, 2001
5
6
7
•Consider A/P calls as
binary response against
log(known frequency)
•Compute sensitivity as
70% likelihood level by
either interpolation or
logistic regression
•“Dampen” computed
frequencies below
sensitivity:
•F < 0: F’ = avg(0,S)
•0<F<S: F’=avg(F,S)
10
How well does it work?
October 11, 2001
11
Reproducibility of F metric (A array)
0h
36h
1
1
AD
ADs F
0
1
MEDACV
F
0.5
October 11, 2001
AD ADs
F
F
Absent
Present
F
AD
1
AD ADs
F
0.5
0
Absent Present
48h
AD ADs
0
MEDACV
0.5
MEDACV
MEDACV
AD ADs
0.5
ADs
Absent Present
60h
AD ADs
F
AD
F
ADs
0
Absent
Present
12
Example of spike-skewed
hybridization (36 hr sample)
2000
1800
Worm Genes
cRNA spikes
•cRNA spikes are
well normalized at
the expense of
worm genes
1600
frequency 36h
1400
1200
• Suggests
inconsistency
between ratio of
spikes to worm
cRNA across
samples: spike
skew
1000
800
600
400
200
0
0
October 11, 2001
500
1000
frequency 36h
1500
2000
13
Sources of spike skew
• Actual concentration of spikes may not
be nominal due to variation in cRNA
“purity”
• Causes: liquid handling of small
microlitre volumes, side reactions in
cDNA/IVT process produce UVabsorbing, non-hybridizable
contaminants
• Result: random per-hybe noise term
introduced into normalized frequencies
October 11, 2001
14
An alternative hybrid
normalization:
Scaled frequency (Fs)
• Need to reduce or eliminate spike skew
as a source of experimental variation in
normalized frequencies
• Average the globally scaled spike
response over a complete set of arrays
October 11, 2001
15
Scaled frequency description
• Define a set of arrays
• Compute ADs for all arrays
• Pool spike responses and fit single
model to pooled response
• Calibrate all arrays with single
calibration factor
• Compute array sensitivity and dampen
frequencies as in the frequency
approach.
October 11, 2001
16
A pooled, scaled spike response
fitted slope: 0.146162419368372
P
P
P
P
P
P
P
P
P
P
P
P
P
P
P
P
P
P
P
P
P
P
P
P
P
P
P
P
P
P
P
P
P
P
P
P
P
3
P
2
log10 average_difference
4
P
P
P
P
P
P
P
P
P
P
P
P
P
P
P
P
P
P
P
P
P
P
P
P
P
P
P
P
P
P
P
P
P
P
P
P
P
P
P
P
P
P
P
•Fit response with
S-plus GLM,
gamma error
model, zero
intercept.
1
P
P
P
P
P
P
P
P
P
P
P
P
P P
P
P
P
P
P
P
P
P
P
P
P
P
P
P
P
P
P
P
P
P
P
P
P
P
P
1
2
3
log10 ppm
October 11, 2001
17
Reproducibility of Fs metric (A array)
October 11, 2001
18
Scaled frequency: cross design
reproducibility (A,B,C arrays)
Three messages
tiled on all array
designs and called
Present on all 0h
arrays
October 11, 2001
19
Conclusions
• Array response to spiked cRNAs can be close to linear over 2.5
logs of concentration.
• A chip sensitivity metric can be computed from Absolute
Decisions associated with spikes; a very useful QC metric.
• Normalization based only on spikes performs inconsistently in
some cases due to ill-quantitation of cRNAs, but can still be
valuable when constant-mean assumption is violated. Better
cRNA quantitation and process control will help.
• A hybrid approach based on global scaling and spikes performs
the same as global AD scaling for single designs, and also
allows cross-design comparisons
October 11, 2001
20
Acknowledgements
•
•
•
•
•
•
Donna Slonim
Maryann Whitley
Yizheng Li
Bill Mounts
Scott Jelinsky
Gene Brown
October 11, 2001
Harvard University:
•Craig Hunter
•Ryan Baugh
21
Extra slides follow ( not part of
presentation)
October 11, 2001
22
Simulations (description)
• Simulations were performed
• Governing equation:
ADij  bij 
October 11, 2001
ADB
i
 a j  mij  sij  rij
23

Figure 4
CV characteristics of simulated data
October 11, 2001
24
Simulations: spike skew degrades reproducibility of frequency
(A array)
October 11, 2001
25
Figure 7
Simulations: spike skew degrades accuracy of frequency
October 11, 2001
26