Quality Control for spotted arrays

April 19, 2007
Agnes Paquet¹, Andrea Barczak¹, (Jean) Yee Hwa Yang²

1. Department of Medicine, Functional Genomics Core Facility, University of California, San Francisco
paquetagnes@yahoo.com
2. School of Mathematics and Statistics, University of Sydney, Australia
http://arrays.ucsf.edu/analysis/arrayquality.html

1. Introduction to arrayQuality

ArrayQuality is a R package, available as part of Bioconductor, designed to help assessing quality of spotted array experiments at several stages of the microarray lifecycle. It provides reports containing several plots and statistical measures that can help you determine if your hybridizations and slides are of good quality. More information about Bioconductor is available at http://www.bioconductor.org.

This guide provides an introduction to microarray quality and a description of the main functionnalities of the package. A full description of the package is given by the individual function help documents available from the R online help system. To access the online help, type help(package=limma) at the R prompt or else start the html help system using help.start() or the Windows drop-down help menu.

2. Installing arrayQuality

Requirements

ArrayQuality is a library for the R project, part of Bioconductor. You will need to have R installed on your computer before installing arrayQuality. For more information about R, see the R project at http://www.r-project.org. ArrayQuality can work on different files at the same time, ONLY if they are from the SAME print-run (same GAL file). If you want to generate quality reports of slides from different print-runs, you need to place them in different folders, one for each print-run.

Installing arrayQuality

ArrayQuality can be installed either from Bioconductor or from the functional Genomics Core Facility web site at http://arrays.ucsf.edu/software. The version from Bioconductor is updated every 6 months. If you would like to use a more recent version, you can obtain the latest one from  http://arrays.ucsf.edu/software or from the developmental version of Bioconductor.

• Automatic installation from Bioconductor
  This is the easiest way to install arrayQuality, as other Bioconductor packages required to use arrayQuality should be automatically installed. Start R on your computer and make sure you are connected to the internet. At the R prompt, type:
  source("http://www.bioconductor.org/biocLite.R")
  biocLite("arrayQuality")
  
  If you are using RGUI (Windows):
  In the R menu, click on "Packages". In the drop-down menu, select "Install package(s) from Bioconductor". Then, select arrayQuality in the list and click "OK".
  
  


• Automatic installation from the Core Facility web site
  Start R on your computer and make sure you are connected to the internet.
  
  You can  install the latest version directly from the Functional Genomics Core Facility web site. At the R prompt, type:
  install.packages("arrayQuality", contriburl="http://arrays.ucsf.edu/software")
  
  If you do not have privileges on your computer to write to the R library directory, for example if you are using a shared unix machine and you are not superuser, you may need to type instead
  install.packages("arrayQuality", contriburl="http://arrays.ucsf.edu/software" lib="myRlibdir")
  
  
• To load arrayQuality:
  Type at the R prompt:
  library(arrayQuality)
  
  
• Manual installation
  Download arrayQuality as a .zip file (Windows users) from Bioconductor or our web site. Start R on your computer and make sure you are connected to the internet.
  In the R console, click on "Packages".
  In the drop-down menu, select "Install package(s) from local zip file..."
  Browse to the zip file you would like to install, and click "Open".

3. Quick starting guide to arrayQuality

General hybridization quality

This component of the package is aimed at verifying the performance of your hybridization, given the good quality of the slide, before any preprocessing steps or further quality assessment on individual spots. Our package provides two kinds of quality control plots:

• A qualitative quality control measurement as a diagnostic plot. It is a quick visual way to determine hybridization quality gathering information from several statistical tools.



• A more quantitative comparison of slide quality as a comparative boxplot. We extract some statistical measures from the test slide and we compare them against results of “good quality” slides to assess the quality of the hybridization.

To generate quality plots from image processing output files

1. Create a directory and move the image processing output files (e.g. .gpr files) of the slides of interest to this directory. Make sure that all files in the directory come from the SAME print-run (same GAL file).



2. Start R, and change R working directory to the one you have just created. In the R menu, select File, then click on “Change dir…”. Browse to your directory from the pop-up window, or enter it manually, and click OK. To double check that you are in the correct directory: in the File menu, click on “Display file(s)…”.



3. To load the package in your R session: type
   library(arrayQuality)
   
   If needed, you may have to install other required packages like marray, limma, convert and hexbin.



4. To generate both diagnostic plots and comparative boxplots on all files in the directory, type:
   result <- gpQuality(organism=”Mm”)



5. To generate diagnostic plots only, run:
   result <- gpQuality(organism="Mm", compBoxplot="FALSE")
   In this case, quantitative quality measures will not be calculated and the HTML report will not be generated.



6. To write down your quantitative quality measures and your normalized data to a file: set output = TRUE when calling gpQuality:
   result <- gpQuality(organism="Mm", output=TRUE)



7. By default, arrayQuality uses print-tip loess normalization. If you prefer to use another method, you can specify it in the norm argument:
   result <- gpQuality(norm="none")
   For more details about normalization methods, please refer to marray package help.

To generate quality plots from marrayRaw or RGList objects

• To generate diagnostic plots: if rawdata is a marrayRaw or RGList object, type in R:
  maQualityPlots(rawdata)

Results

• In the working directory: - two plots for each test slide (a diagnostic plot and a comparative boxplot) - a HTML quality report



• In the R global environment, the function returns a list of two objects: - A marrayRaw object describing all tested slides - A quality measures matrix: this matrix contains all comparison measures values extracted for each test slide. Each column of the matrix represents a different slide.

Print-run quality control

This component of arrayQuality provides diagnostic plots for 9-mers hybridization and Quality Control hybridization.

9-mers analysis: PRv9mers()

To generate diagnostic plots

1. Copy all 9-mers hybridizations gpr files from the SAME print-run (same GAL file) to a directory.



2. Start R and change R working directory to the one containing your gpr files (see above for more details).



3. Load the package in your R session:
   library(arrayQuality)
   
   If needed, you may have to install other required packages like marray, limma, convert and hexbin.



4. To generate diagnostic plots, type:
   PRv9mers(prname=”12Mm”)
   The prname argument represents the name of your print-run. For more details about other arguments, please refer to the online manual.

Results

1. Diagnostic plots as image in .png format for each tested slide
2. An Excel file (typically named 9Mm9mer.xls, where 9Mm is the name of your print-run, as passed to prname) containing for each spot on the slide:
   • Name and ID of the spot
   • The probability of being present or absent (p from EM algorithm). If several files are tested together, you will have a probability of being present/absent for each file. A spot is considered absent if p < 0.5.
   • The average probability of being present or absent.
   • The raw signal intensity (Signal column) or average raw signal intensity if several files are tested together for each spot.
3. An Excel file (typically named 9MmMissing.xls, where 9Mm is the name of your print-run, as passed to prname) containing information on missing probes only:
   • Name and ID of the spot
   • The probability of being present or absent (p from EM algorithm). If several files are tested together, you will have a probability of being present/absent for each file. A spot is considered absent if p < 0.5.
   • The average probability of being present or absent.
   • The raw signal intensity (Signal column) or average raw signal intensity if several files are tested together for each spot.
4. A text file (typically named 9MmQuickList.txt, where 9Mm is the name of your print-run, as passed to prname) containing the missing probes ids, each on a separate line. This file can be opened in any word processing programd as well as being a "quick list" in Acuity software (http://www.axon.com/gn_Acuity.html).

Quality Control hybridization: PRvQCHyb()

1. Detect any difference in overall signal intensity compared to other print-runs
   • 70-mers oligonucleotides hybridizations
   • Selection of several test slides to ensure that the same quantity of material was spotted across the platter, as a print-run will generate 255 slides using the same well for one probe. QCHybs are performed using one slide from the beginning of the print, one from the middle, one from the end (e.g. numbers 20,100 and 255 in the Functional Genomics Core Facility).
2. Check if the GAL file was generated properly, i.e. check that no error was made with ordering or orientation of the plates during the print.
3. Reproducibility:
   A good way to verify the quality of a new print is to hybridize known samples to new slides. Then, we can compare signal intensity from the new slides to existing data, and check that there is no loss in signal. Log ratios (M) for known samples should be similar across print-runs. Example of samples used for QCHybs includes Human Reference pool, Mouse liver, Mouse lung, with dye swaps.

To generate diagnostic plots

1. Create a directory and move the image processing output files (.gpr files only) of the slides of interest to this directory. Make sure that all files in the directory come from the SAME print-run (same GAL file).



2. Start R, and change R working directory to the one you have created (see general hybridization paragraph above for more details)



3. Load the package in your R session:
   library(arrayQuality)
   
   If needed, you may have to install other required packages like marray, limma, convert and hexbin.



4. For QCHyb analysis, run the following command:
   
   PRvQCHyb()
   
   

Results

Introduction to microarray quality

A microarray experiment is composed of several steps, including experimental design, sample preparation, and various statistical analyses (figure 1). They are represented in the microarray lifecycle below. As microarray technology is complex and sensitive, it is important to assess the performance of each step before going to the next one. In addition, this is also a good way to trace back the cycle to understand potential causes for upstream problems.

Figure 1: Microarray experiment lifecycle

1. Print quality
   
   • 9mers hybridization
   • Quality Control hybridization
2. mRNA quality
3. Array hybridization quality
4. Spot quality

Figure 2: Quality Control for spotted arrays experiment

1. Print quality:
   
   This component is highly tailored to the Shared Genomics Core Facility at UCSF, but the framework can be adapted to other Core facilities or laboratories printing their arrays. It is an essential component of a printed array experiment, as any print pin, probe or slide surface defect will affect the quality of hybridization to the slide, and this can’t be fixed by statistics. Only prints that did pass the quality control check will be used for actual hybridization.



2. Hybridization quality:
   
   This is a global assessment of the hybridization performance. It helps determine for example any problem with the dyes, or uneven hybridization. Then, once you have determined that your hybridization is good, you can look at each individual spot quality, remove bad spots, and perform statistical analysis.

3. Print-run quality control

When a print-run is completed, it is necessary to verify the quality of the resulting arrays. This can be done by using two kinds of hybridization to the new slides. The first type of hybridization, which we term “9mers hyb”, uses small oligonucleotides (random 9-mers), which will hybridize to each probe. This hybridization will help to determine the quality of spot morphology as well as the presence or absence of spotted oligonucleotides. The resulting data will be used to create a list of all missing spots.

The second type of hybridization, which we will term Quality Control Hybridization (QCHyb), uses mRNA from predefined cell lines (e.g. liver vs. pool, K562 vs. Human Universal Reference pool from Stratagene). These hybridizations can be use as a more quantitative description of the slides. The same comparison hybridizations are done for different print-run, assessing their reproducibility. QCHybs are also used to verify accuracy of GAL files, number of missing spots, binding capacity, background signal intensity…

The arrayQuality package provides specific tools to help assess quality of slides for both 9-mers and QC hybridization.

3.1 9-mers hybridizations

• Copy all 9-mers hybridizations gpr files from the SAME print-run (same GAL file) to a directory.
• Change R working directory to the one containing your gpr files as described in section 1.
• Type:
  PRv9mers(prname=”12Mm”).

Results

1. Diagnostic plots as image in .png format for each tested slide



2. An Excel file (typically named 9Mm9mer.xls, where 9Mm is the name of your print-run, as passed to prname) containing for each spot on the slide:
   
   • Name and ID of the spot
   • The probability of being present or absent (p from EM algorithm). If several files are tested together, you will have a probability of being present/absent for each file.
     A spot is considered absent if p < 0.5.
   • The average probability of being present or absent.
   • The raw signal intensity (Signal column) or average raw signal intensity if several files are tested together for each spot.
3. An Excel file (typically named 9MmMissing.xls, where 9Mm is the name of your print-run, as passed to prname) containing information on missing probes only:
   
   
   • Name and ID of the spot
   • The probability of being present or absent (p from EM algorithm). If several files are tested together, you will have a probability of being present/absent for each file.
     A spot is considered absent if p < 0.5.
   • The average probability of being present or absent.
   • The raw signal intensity (Signal column) or average raw signal intensity if several files are tested together for each spot.



4. A text file (typically named 9MmQuickList.txt, where 9Mm is the name of your print-run, as passed to prname) containing the missing probes ids, each on a separate line. This file can be opened in any word processing program.

Description of the diagnostic plots

1. The first column (left) represents boxplots of log intensity, by plates (top) and by print-tip group (bottom). In this example, you will notice on the boxplot by plates (top left corner) that plates 44 and 48 have lower intensity and wider range than the others. Both plates contain mostly empty controls, as designed by Operon.
2. Central plot: spatial plot of intensity. This helps to locate missing spots. The color scale reflects the signal intensity, the darker the color of the plot, the stronger the signal. Missing spots are represented in white. In Figure 3 spatial plot, top right corner white spots come from the empty spots.
3. Right column: Density plot of the foreground and background log intensity.
   1. Foreground density plot: it should be composed of 2 peaks. A smaller peak in the low intensity region containing missing spots and negative control spots, and a higher one representing the rest of the spots (probes). The number of present and absent spots, excluding empty controls, estimated by EM algorithm is indicated on the graph.
   2. Background density plot: one peak in the low intensity region. If a slide is of good quality, the background peak should not overlap too much with the foreground peak corresponding to the bulk of the data.

Examples

Figure 3: Example of diagnostic plot for 9-mers hybridization

3.2 Quality Control hybridizations 

9-mers hybridizations help verify that oligonucleotides have been spotted properly on the slides. The next print-run quality control step will be:

1.      Detect any difference in overall signal intensity compared to other print-runs

a.      70-mers oligonucleotides hybridizations

b.      Selection of several test slides to ensure that the same quantity of material was spotted across the platter, as a print-run will generate 255 slides using the same well for one probe. QCHybs are performed using one slide from the beginning of the print, one from the middle, one from the end (e.g. numbers 20,100 and 255 in the Functional Genomics Core Facility).

2.      Check if the GAL file was generated properly, i.e. check that no error was made with ordering or orientation of the plates during the print.

3.      Reproducibility:

A good way to verify the quality of a new print is to hybridize known samples to new slides. Then, we can compare signal intensity from the new slides to existing data, and check that there is no loss in signal. Log ratios (M) for known samples should be similar across print-runs. Example of samples used for QCHybs includes Human Reference pool, Mouse liver, Mouse lung, with dye swaps.

The function in the package which performs the quality assessment for QCHybs is PRvQCHyb().

-          Copy the QCHybs gpr files from the SAME print-run (same GAL file) in a directory.

-          Change R working directory to the one containing your gpr files as described in section 1.

-          Type:

> PRvQCHyb(prname="9Mm")

where prname is the name of the print-run. For more details about its arguments, please refer to the online manual.
 

Results 

PRvQCHyb() returns a diagnostic plot as an image in .png format for each tested slide.

Throughout our document, we will be using the color code described in Table 1 to highlight control spots.

Table 1: Color code used in arrayQuality

Restrictions:

Currently, PRQCHyb() supports Mouse genome (Mm) only. We will add Human data as soon as it becomes available.

Description of the diagnostic plots

Figure 4 shows an example of a nice print-run QCHyb.

1. MA-plot of raw M values. No background subtraction is performed. The colored lines represent the loess curves for each print-tip group. The red dots highlight  any spot with corresponding weighted value  less than 0. Users can create their own weigthing scheme or function. Things to look for in a MA-plot are saturation of spots and the trend of loess curves, which is an indicator of the amount of normalization to be performed.
   

2. Boxplot of raw M values by print-tip group, without background subtraction.

3. Spatial plot of rank of raw M values (no background subtraction): Each spot is ranked according to its M value. We use a blue to yellow color scale,where blue represents the higher rank (1), and yellow represents the lower one. Missing spots are represented as white squares.

4. Spatial plot of A values. The color indicates the strength of the signal intensity, i.e. the darker the color, the stronger the signal. Missing spots are represented in white.

5. Histogram of the signal-to-noise log-ratio (SNR) for Cy5 and Cy3 channels. The mean and the variance of the signal are printed on top of the histogram. In addition, overlay density of SNR stratified by different control types (status) are highlighted. Their color schemes are provided in Table 1. The SNR is a good indicator for dye problems. The negative controls and empty controls density lines should be closer, almost superimposed.

6. Comparison of Mvalues of probes known to be differentially expressed from the tested array to average Mvalues obtained during previous hybridizations. This plot is aimed at verifying the reproducibility of print-runs. The dotted lines are the diagonal (no change) and the +2/-2 fold change lines. Each probe is represented by a number, and described in the file MmDEGenes.xls. Most of the spots should lie between the +2/-2 fold-change regions. If the technique was perfect, you should see a straight line on the diagonal. If any probe falls off this region (number 29 here), you can look up its number in our probe list in MmDEgenes.xls and get more information about it.

7. Dot plot of controls A values, without background subtraction. Controls with more than 3 replicates are represented on the Y-axis, the color scheme is represented in Table 1. Intensity of positive controls should be in the high-intensity region, negative and empty controls should be in the lower intensity region. Positive controls range and negative/empty controls range should be separated. Replicate spots signal should be tight.
   

Example

Data for this example was provided by the Functional Genomics Core Facility in UCSF. We have tested slide number 137 from print-run 9Mm. This print-run uses Operon Version 2 Mouse oligos. Results are represented figure 4.
 

> library(arrayQuality)

> datadir <- system.file("gprQCData", package="arrayQuality")

> PRvQCHyb(fnames=”9Mm137.gpr”, path=datadir, prname="9Mm")

Figure 4: Diagnostic plot for print-run Quality Control hybridization

4. General hybridization quality

This component is aimed at verifying the performance of your hybridization, given the good quality of the slide, before any preprocessing steps or further quality assessment on individual spots. This is where you determine if your experiment quality is good enough to enter your dataset. For example, you will need to remove any hybridization with very low SNR, or large spatial artifacts.

Our package provides two kinds of quality control plots. The first one is a qualitative quality control measurement as a diagnostic plot. It is a quick visual way to determine hybridization quality gathering information from several statistical tools. More details on individual diagnostic plots can be found in the vignette “marrayPlots” in the package marray. The second one is a more quantitative comparison of slide quality. We extract some statistical measures from the test slide and we compare them against results obtained for a collection of slides of “good quality” to assess the quality of the hybridization. This comparison is visualized through a comparative boxplot. Results are displayed in a HTML report. Figure 5 shows a screen shot of a typical HTML report. Users can click on each image to obtain a higher resolution plot.

Diagnostic plots can be generated  for different image processing software format: GenePix format files (.gpr files), Spot format files (.spot) and Agilent format files, or from marrayRaw or RGList objects. Most arguments can also be customized to match your own data: which probes are used as controls, which column of the image processing output file is used to define your spot types... You can also specify your own collection of good quality slides using the functions globalQuality and qualRefTable. For more details about these functions, please refer to the online help and the example at the end of this Section.

To generate quality plots: gpr files: gpQuality()

-          Copy the gpr files from the SAME print-run (same GAL file) in a directory.

-          Change R working directory to the one containing your gpr files as described in Section 1

-          To generate both diagnostic plots and comparative boxplots on all files in the directory, run:
       
         > result <- gpQuality(organism=”Mm”)

         -          To generate diagnostic plots only, run:
                
         >  result <- gpQuality(organism="Mm", compBoxplot="FALSE")

In this case, quantitative quality measures will not be calculated and the HTML report will not be generated.

- To write down your quantitative quality measures and your normalized data to a file: set output = TRUE when calling gpQuality:

To generate quality plots from marrayRaw/RGList objects: maQualityPlots

-          To generate diagnostic plots: if rawdata is your marrayRaw/RGList object, type:

> maQualityPlots(rawdata)

Results

gpQuality() outputs

• In the working directory:

-          two plots for each test slide (a diagnostic plot and a comparative boxplot)

-          a HTML quality report

• In the R global environment, the function returns a list of two objects: 
  

        -    A marrayRaw object describing all tested slides

        -    A quality measures matrix: this matrix contains all comparison measures values extracted for each test slide. Each column of the matrix represents a different slide.

For each slide, you will find on the report how many of your slide’s results are below the recommended range. If you want to specify a directory to store the results, you can do it by modifying the argument resdir accordingly. For more details about gpQuality arguments, please refer to the online manual.

Figure 5: Example of HTML report generated by gpQuality

• In the working directory:

Restrictions

gpQuality calls two key functions, maQualityPlots and qualBoxplot. qualBoxplot supports Mouse (Mm) and Human (Hs) genomes only. To generate quality plots for other genomes, you need to set gpQuality argument compBoxplot = FALSE. In this case, only the diagnostic plots will be generated.

Description of the diagnostic plots:

Figure 6 represents an example of a good hybridization diagnostic plot.

1. MA-plot of raw M. No background subtraction is performed. The colored lines represent the loess curves for each print-tip group. The red dots highlight  any spot with corresponding weighted value  less than 0. Users can create their own weigthing scheme or function. Things to look for in a MA-plot are saturation of spots and the trend of loess curves, which is an indicator of the amount of normalization to be performed.

2. MA-plot of normalized data density. By default, print-tip loess normalization is used. Instead of the typical MA-plot, we have used the package "hexbin" to highlight density of dots on the MA-plot. A light yellow color indicates a high density of dots, whereas blue color represents a lower density. This plot gives you information on the bulk of your data intensity (low/high signal)

3. Spatial plot of rank of raw M values (no background subtraction): Each spot is ranked according to its M value. We use a blue to yellow color scale,where blue represents the higher rank (1), and yellow represents the lower one. Missing spots are represented as white squares. This is a quick way to visually detect uneven hybridization and missing spots.

4. Spatial plot of normalized M values ranks. By default, print-tip loess normalization is used. Each spot is ranked according to its M value. We use a blue to yellow color scale,where blue represents the higher rank (1), and yellow represents the lower one. Missing spots are represented as white squares. In addition, flagged spots are higllighted by a black square. This type of graphical representation helps verify that normalization removed any spatial effects.

5. Spatial plot of raw A values. The color indicates the strength of the signal intensity, i.e. the darker the color, the stronger the signal. Missing spots are represented in white.

6. Histogram of the signal-to-noise log-ratio (SNR) for Cy5 and Cy3 channels. The mean and the variance of the signal are printed on top of the histogram. In addition, overlay density of SNR stratified by different control types (status) are highlighted. Their color schemes are provided in Table 1. The SNR is a good indicator for dye problems. The negative and empty controls density lines should be closer, almost superimposed.

7. Dot plot of controls normalized M values. Controls with more than 3 replicates are represented on the Y-axis, the color scheme is represented in Table 1. Controls M values should be tight. and close to 0.
   

8. Dot plot of controls A values, without background subtraction. Controls with more than 3 replicates are represented on the Y-axis, the color scheme is represented in Table 1. Intensity of positive controls should be in the high-intensity region, negative and empty controls should be in the lower intensity region. Positive controls range and negative/empty controls range should be separated.
   

Description of the comparative Boxplot:

Figure 7 shows an example of a comparative boxplot.

We have chosen a wide range of measures to quantify the quality of a typical hybridization: single channel measures (range of foreground signal, MAD of background, signal to noise ratio…), two channel measures (median A values for each type of controls, amount of normalization needed…), percentage of flagged spots... Some measures have been negated such that the quality scale had an increasing trend from problematic to good quality.

For each measure, we have represented the following on the graph :

-         Boxplot of the reference slides values.

-        1^st and 3^rd quantiles before scaling for each boxplot.

-         Y-axis on the right : for each measure, we have printed 2 values. The first one is the percentage of reference slides measures under your slide’s result. The second one is your slide value for this measure before scaling.

1. rangeRf: Range of Cy5 foreground, where the range is defined by:

rangeRf = max(log2 (median Cy5 foreground)) - min(log2(median Cy5 foreground))
where median Cy5 foreground corresponds to the "F635 Median" column of the gpr file.

2. rangeGf: Range of Cy3 foreground, where the range is defined by:

rangeGf = max(log2 (median Cy3 foreground)) - min(log2(median Cy3 foreground))
where median Cy3 foreground corresponds to the "F532 Median" column of the gpr file.

3. -RbMad: Cy5 background MAD

RbMad = mad[log2(Cy5 background)]

where:
- Cy5 background corresponds to the "B635 Median" column of the gpr file
- MAD = median{ | Y –mu | }, when Y is normal

4.  -GbMad: Cy3 background MAD

GbMad = mad[log2(Cy3 background)]

7. -Median A for empty control:

Median A = [ log2(median Cy5 foreground) + log2(Cy3 foreground) ] / 2

Median A for empty control = median( A(Empty controls))

where:
- median Cy5 foreground corresponds to the "F635 Median" column of the gpr file
- median Cy3 foreground corresponds to the "F532 Median" column of the gpr file
- Empty controls are the probes labelled "Empty"

8. -Median A for negative control:

Median A = [ log2(median Cy5 foreground) + log2(Cy3 foreground) ] / 2

Median A for negative control = median( A(Negativecontrols))

where:
- median Cy5 foreground corresponds to the "F635 Median" column of the gpr file
- median Cy3 foreground corresponds to the "F532 Median" column of the gpr file
- Negative controls are the probes labelled "Negative"

9. Median A values for Positive controls: 

A = [ log2(median Cy5 foreground) + log2(Cy3 foreground) ] / 2

Median A for positive control = median( A(Positive controls))

where:
- median Cy5 foreground corresponds to the "F635 Median" column of the gpr file
- median Cy3 foreground corresponds to the "F532 Median" column of the gpr file
- Positive controls are the probes labelled "Positive"

10. Difference between A values for Positive controls and  A values for Negative controls

difference =  median( A(Positive controls)) - median( A(Negativecontrols))

11. -varRepA: variance of replicates spots A values

varRepA = var[ A(replicates) ]

where:
- A = [ log2(median Cy5 foreground) + log2(Cy3 foreground) ] / 2
- median Cy5 foreground corresponds to the "F635 Median" column of the gpr file
- median Cy3 foreground corresponds to the "F532 Median" column of the gpr file

12. -msePtip: MSE of M values by print-tip group, no background subtraction

 M =  log2(Median Cy5 foreground) - log2(Median Cy3 foreground)
msePtip = MSE( mean M by print-tip)

where:
- median Cy5 foreground corresponds to the "F635 Median" column of the gpr file
- median Cy3 foreground corresponds to the "F532 Median" column of the gpr file
- MSE(X) = E( (X-t)² ), with t a parameter and X an estimator of t.
 

13. -mseFit: MSE of lowess curve

fit = lowess(A, M) 
mseFit = MSE(fit$y)

where:
- A = [ log2(median Cy5 foreground) + log2(Cy3 foreground) ] / 2
- M =  log2(Median Cy5 foreground) - log2(Median Cy3 foreground)
- median Cy5 foreground corresponds to the "F635 Median" column of the gpr file
- median Cy3 foreground corresponds to the "F532 Median" column of the gpr file
- MSE(X) = E( (X-t)² ), with t a parameter and X an estimator of t.

14. -Percentage of flagged spots

[number of spot with flag < 0 / number of spots] * 100

where flag is the information from the "Flags" column of the gpr file. Only spots with flag less than 0 are taken into account.

 
15. -M values MMRmad

MMR = Mmean – Mmedian

MAD(MMR)

where:
- Mmean =  log2(Mean Cy5 foreground) - log2(Mean Cy3 foreground)
M values calculated using mean signal

- Mmedian = log2(Median Cy5 foreground) - log2(Median Cy3 foreground)
M values calculated using median signal

- mean Cy5 (Cy3) foreground is the "F635 Mean" ("F532 Mean") column of the gpr file
- median Cy5 (Cy3) foreground is the "F635 Median" ("F532 Median")  column of the gpr file
- MAD = median{ | Y –mu | }, when Y is normal
 

16. -Percentage of spots with abs[MMR] > 0.5

where:
- MMR = Mmean – Mmedian

- Mmean =  log2(Mean Cy5 foreground) - log2(Mean Cy3 foreground)
M values calculated using mean signal

- Mmedian = log2(Median Cy5 foreground) - log2(Median Cy3 foreground)
M values calculated using median signal

Example

Data for this example was provided by the Functional Genomics Core Facility in UCSF. We have tested slide number "137" from print-run "9Mm". This array was fabricated using Operon Version 2 Mouse oligos and the hybridization measures differential gene expression in two RNA samples, Mouse Liver and Mouse Reference Pool. Results are represented Figure 5 and Figure 6.

To generate diagnostic plots, comparative boxplots, HTML report and to write your quality measure and normalized data to a file in a directory named "Results":

> library(arrayQuality)

> datadir <- system.file("gprQCData", package="arrayQuality")

> result <- gpQuality(fnames = "9Mm137.gpr", path = datadir,organism = ”Mm”, output = TRUE, resdir = "Results")

 
Figure 6: General hybridization quality diagnostic plot

Figure 7: Comparative boxplot

Customizing gpQuality

> controlCode <- readcontrolCode(file=”mySpotTypes.txt”, controlId="ID")

- Find which column of the gpr file can be used to identify your new spot types. It is typically the "ID" or the "Name" column.
- To generate both types of plots: call gpQuality specifying your new controlCode matrix in controlMatrix and which column is used to define your spot types in controlId.

> result <- gpQuality(controlMatrix = controlCode, controlId=”Id”)