STAT530lecture5Oligoarray - Ping Ma STAT 530 1 STAT 530...

Info iconThis preview shows pages 1–7. Sign up to view the full content.

View Full Document Right Arrow Icon

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
This is the end of the preview. Sign up to access the rest of the document.

Unformatted text preview: Ping Ma STAT 530 1 STAT 530 STAT 530 Oligonucleotide Oligonucleotide Microarray Microarray Ping Ma Ping Ma 2 Ping Ma STAT530 Source: Affymetrix, Inc. Ping Ma STAT 530 2 3 Ping Ma STAT530 4 Ping Ma STAT530 Affymetrix Microarray Imagine Analysis s Affymetrix GeneChip Operating System (GCOS) cel file X Y MEAN STDV NPIXELS 701 523 311.0 76.5 16 702 523 48.0 10.5 16 cdf file Which X,Y correspond to which probe sequence, and targeted transcript Ping Ma STAT 530 3 5 Ping Ma STAT530 Normalization Normalization s Between Between-array normalization array normalization------ Cyclic Loess Cyclic Loess 1. Rank 1. Rank-invariant set invariant set 2. Loess 2. Loess ------ Baseline Normalization ( Baseline Normalization ( dChip dChip )------ Quantile Quantile Normalization (RMA) Normalization (RMA) 6 Ping Ma STAT530 Before Normalization s Pairwise MA plot for 5 arrays, probe (PM) 2 2 log ( / ) log i j i j M PM PM A PM PM = = Ping Ma STAT 530 4 7 Ping Ma STAT530 After Normalization s Pairwise MA plot for 5 arrays, probe (PM) 2 2 log ( / ) log i j i j M PM PM A PM PM = = 8 Ping Ma STAT530 Modeling probe effects Modeling probe effects s 1) Probes sequence has different hybridization efficiency s 2) cross hybridization, SNP, alternative splicing s 3) Probe position effect, 3 bias s Probe effects can dominate biological variation of interest Ping Ma STAT 530 5 9 Ping Ma STAT530 Model-Based Expression Index s Each probe signal is proportional to Amount of target sample: Affinity of specific probe sequence to the target: j s High affinity probe not necessarily better 10 Ping Ma STAT530 Model for One Gene Model for One Gene 1 1 1 and : observed MM and PM values for chip i, probe j, :baseline hybridization level for probe pair j, : expression index in chip i, , : rate fa ij j i j ij ij j i j i j ij ij ij j i j j X X X X = + + = + + + ctors characterizing probe pair j constraint: 0, 0, 0, j i j j Ping Ma STAT 530 6 11 Ping Ma STAT530 Using PM/MM Differences Using PM/MM Differences s PM/MM differences eliminate most background PM/MM differences eliminate most background and cross and cross-hybridization signals hybridization signals s Affyemtrix Affyemtrix s GeneChip GeneChip software is using average software is using average differences as basis for determining fold differences as basis for determining fold changes, and their validation showed average changes, and their validation showed average differences are linear to mRNA concentrations differences are linear to mRNA concentrations at certain dynamic range at certain dynamic range...
View Full Document

This note was uploaded on 11/21/2010 for the course STAT STAT530 taught by Professor Ma during the Spring '07 term at University of Illinois, Urbana Champaign.

Page1 / 14

STAT530lecture5Oligoarray - Ping Ma STAT 530 1 STAT 530...

This preview shows document pages 1 - 7. Sign up to view the full document.

View Full Document Right Arrow Icon
Ask a homework question - tutors are online