Modelbased analysis of oligonucleotide arrays d Chip software
Model-based analysis of oligonucleotide arrays, d. Chip software Cheng Li (Joint work with Wing Wong) Statistics and Genomics – Lecture 4 Department of Biostatistics Harvard School of Public Health January 23 -25, 2002
Source: Affymetrix website
Custom software: raw image
Custom software: getting representative value of a probe cell
Normalization is needed to minimize non-biological variation between arrays
Normalization methods • Current software uses linear normalization • Nonlinear curve fitting based on scatter plot is still inadequate because 1) effects of differentially expressed genes may be “normalized” 2) regression phenomenon and asymmetry
Regression phenomenon and asymmetry
Invariant set normalization method • A set of points (xi, yi) is said to be order-preserving if yi < yj whenever xi < xj • The maximal order-preserving subset can be obtained by dynamic programming • If a gene is really differentially expressed, it’s cells tend not to be included into an large orderpreserving subset • Our method is based on an approximately order preserving subset, called “Invariant set”
Fig. 2. 9 Normalization of a pair of replicated arrays
Figure 2. 10. Two different samples. The smoothing spline in (A) is affected by several points at the lower-right corner, which might belong to differentially expressed genes. Whereas the “invariant set” does not include these points when determining normalization curve, leading to a different normalization relationship at the high end.
A pair of split-sample replicate arrays
Source: Affymetrix website
Data for one probe set, one array PM/MM differences eliminate background and crosshybridization signals
Validation experiments suggest Average Differences are linear to m. RNA concentrations at certain dynamic range Lockhart et al. (1996) Nature Genetics, Vol 14: 1675 -1680
Data for one gene in many arrays
Box plot showing array and probe effects
Modeling probe effects 1) Probes sequence has different hybridization efficiency 2) cross hybridization, SNP, alternative splicing 3) Probe position effect, 3’ bias Probe effects can dominate biological variation of interest Previous method : use multiple probes, average to reduce “noise” Our methods: statistical models for probe effects, “metaanalysis”, learning algorithms, estimation of expression level conditional on knowledge of probe effect
Principal component analysis (42 points in 20 -space) suggests the data matrix has approx. rank 1
Model for one gene in multiple arrays
Figure 1. 1. Black curves are the PM and MM data of gene A in the first 6 arrays. Light curves are the fitted values to model (1). Probe pairs are labeled 1 to 20 on the horizontal axis.
Using PM/MM Differences • PM/MM differences eliminate most background and cross-hybridization signals • Affyemtrix’s Gene. Chip software is using average differences as basis for determining fold changes, and their validation showed average differences are linear to m. RNA concentrations at certain dynamic range
Model for PM/MM differences (1. 2)
Figure 1. 2. Black curves are the PM-MM difference data of gene A in the first 6 arrays. Light curves are the fitted values to model (2).
Residuals of the fitting
Model fitting amounts to fixing ’s and regress to estimate
Fig 1. 5 Array outlier: large standard errors of 4
Fig. 1. 6 Probe outlier: large standard errors of 17 Also see gene 6898
Fig. 1. 4 Array outlier image shows that the model automatically handles image contamination
Compare Model-based expression with Average Difference • The array set 5 has 29 pair of arrays replicated at split-m. RNA level • The differences between the replicated arrays provides a opportunity to assess different expression calculation method
Figure 2. 5. Log (base 10) expression indexes of a pair of replicate arrays (array 1 and 2 of array set 5) for MBEI method (A) and AD method (B). The center line is y=x, and the flanking lines indicate the difference of a factor of two.
(A) (B) Figure 2. 6. Boxplots of average absolute log (base 10) ratios between replicate arrays stratified by presence proportion for (A) MBEI method, (B) AD method.
Source: Affymetrix website
Finding Confidence Interval of Fold Change
Table 2. 1 Using expression levels and associated standard errors to determine confidence intervals of fold changes
Resampling hierarchical clustering using standard errors of model-based expression
Incorporate biological knowledge and database when analyzing microarray data Right picture: Gene Ontology: tool for the unification of biology, Nature Genetics, 25, p 25
Functional significant clusters Found 13 structural protein genes out of a 49 -cluster (all: 198/2622, PValue: 1. 00 e+000)
Statistical analysis of high-density oligonucleotide arrays: a multiplicative noise model R. Sasik and J. Corbeil (UCSF) Problems with LWR model: • The expression index can still be negative. Slides prepared by Xuemin Fang • Genes with negative index can still be classified as present.
Statistical model: • Based on the same assumption as the LW model, that PM intensity is directly proportional to the concentration ci of the transcript, . Write the relation in the form • Our model is • where • Least squared estimation of the parameters. • Constraint:
Algorithm -- When analyzing a batch of ns samples: • Normalize all samples to the first one on the list by requiring the sum of all PM intensities be the same as that of the first sample. • Select the background probes using Naef’s method (MM is used in this step). • Subtract the median of the background probe intensity from every PM probe in the array. • Probes that become negative are eliminated. • Fit the model and probes contributes most to the sum of squares are eliminated. • Normalize again and repeat 1 -5, until the distribution of residuals is Gaussian.
Bias, variance and fit for three measures of expression: Av. Diff, Li & Wong's, Av. Log (PM -bg) Rafael Irizarry, Terry Speed (Johns Hopkins) Slides prepared by Xuemin Fang
A background plus signal model: • Here represents background signal caused by optical noise and non-specific binding. • The mean background level is represented with the random component with. and • The transcript signal contains a probe affinity effect , the log expression measures , and an error term. • Both error terms standard normal. and are independent
Expression index: • A naïve estimate of with is given by the mode of the log 2(MM) distribution. • An estimate of this distribution is obtained using a density kernel estimate.
Acknowledgement Data source: Stan Nelson (UCLA) Sven de Vos (UCLA) Dan Tang (DFCI) Andy Bhattacharjee (DFCI) Richardson Andresa (DFCI) Allen Fienberg (Rockefeller)
- Slides: 45