WHAT IS CUSTOMARRAY™?
CustomArray™ is a unique system for the in situ synthesis of
oligonucleotide microarrays made to your specifications. You design the
CustomArrays using web-based free software; we ship them to you within
30 business days.
HOW DOES CUSTOMARRAY
WORK?
The custom oligonucleotide microarray is synthesized on a semiconductor
using an electrochemical synthesis process. Each oligonucleotide probe
(spot) is synthesized on a platinum electrode that is independently
controlled by the synthesizer’s computer. Synthesis is based on
established phosphoramidite chemistry and occurs at thousands of sites
simultaneously according to a computer algorithm that activates only
specified electrodes. Since physical photolithographic masks,
mechanical micromirrors, or inkjet technologies are not involved in the
process, all probes can be easily changed without extra time or cost.
The oligonucleotides are synthesized in a
proprietary porous reaction layer that coats the surface of the
semiconductor, and acts to covalently attach the oligonucleotide to the
surface of the chip. The resultant CustomArray chip is shown in Fig.1.
Figure 1. A general scheme of CustomArray 12K chips.
HOW IS CUSTOMARRAY
DIFFERENT FROM SPOTTED ARRAYS?
CustomArray is an in situ synthesis technology. Oligonucleotide probes
are synthesized directly on the chip. This means there are no up front
costs of buying hundreds of different oligonucleotides, and making
changes to chip designs and layouts is easy.
HOW IS CUSTOMARRAY
DIFFERENT FROM OTHER IN SITU SYNTHESIS TECHNOLOGIES?
CustomArray chips use electrochemical detritylation to control DNA
synthesis. Electrochemical detritylation is a much more efficient
method to synthesize oligonucleotides than light-based synthesis
processes. This means the oligonucleotides are of the highest quality
and the sensitivity of the microarray is maximized. Since physical
photolithographic masks, mechanical micromirrors, or inkjet
technologies are NOT involved in the process, all probes can be easily
changed without extra time or costs.
WHAT CAN I USE CUSTOMARRAY FOR?
CustomArray can be used to order whole microarrays just like ordering
oligos!
Initially, our platform has been tailored for
gene expression studies. At present it is widely used for other
applications, including comparative genome hybridization (CGH), SNP
analysis and detection, re-sequencing, and chromatin
immunoprecipitation (ChIP) on chip. The web-based system for designing chips
can create optimized probes based on a set of customer-defined
sequences that are submitted as GenBank accession numbers or raw
sequence data. It can also accept customer-specified oligonucleotide
probe sequences.
Our Design-on-Demand™
service can help to change the probe design algorithm to
tailor probes for your particular application.
HOW LONG DOES IT TAKE TO MAKE MY
OWN CUSTOMARRAY?
After you have finalized your array design, we typically ship the
finished arrays within 30 days from receipt of your order.
CAN CUSTOMARRAY™ BE RE-USED AFTER
THE FIRST HYBRIDIZATION?
A single CustomArray™ microarray can be stripped three times using the
CombiMatrix CustomArray™ Stripping Kit. The CustomArray™ Stripping
protocol is based on chemical denaturation of DNA:DNA and DNA:RNA
hybrids between oligonucleotide probes on microarrays and corresponding
labeled targets. As a result of this denaturation, labeled targets are
removed from oligonucleotide probes and washed off microarrays. The
resultant stripped CustomArray™ microarrays can be used in the same way
as newly synthesized ones. In total, the CombiMatrix Stripping Kit
would enable you to use a single CustomArray™ microarray four times.
WHAT
METHOD IS USED FOR QUALITY CONTROL OF CUSTOMARRAY CHIPS?
Each and every synthesized CustomArray chip is quality tested by
hybridization with Cy5-labeled random 9-mer oligonucleotide targets.
This hybridization is used to visualize the synthesis at every
electrode as well as irregular spot morphology. All hybridization data
are first analyzed automatically using rigorous statistical criteria,
and then inspected visually by an independent reviewer for the final
‘pass’ or ‘fail’ decision to ensure the high quality of chips shipped
to our customers. The arrays that have passed the inspection are
stripped of the hybridized 9-mers, washed, and dried for shipment.
HOW
SENSITIVE ARE CUSTOMARRAYS?
We hybridized CustomArrays with a complex labeled target mixture
containing four different spiking control RNA’s added at concentrations
of 0.375, 0.75, 1.5, and 3 pM. The resultant data showed reproducible,
above-background signals for the lowest spike concentration (0.375 pM),
linearity at the applied concentration range (Fig. 2), and low
coefficients of variation among replicate probes within the same array
(on average from 6 to 10%).
Figure
2. Hybridization of spiking controls to 2 CustomArray chips in the
context of a complex labeled target mixture. Four different spiking
control RNAs were added at concentrations of 0.375, 0.75, 1.5, and 3
pM. The average standard deviation across the set of replicate probes
is shown by the error bars.
HOW REPRODUCIBLE ARE
CUSTOMARRAY CHIPS?
We hybridized 20 identical CustomArray chips with the same RNA target
sample prepared from the Universal Human Reference RNA (Stratagene) and
labeled with Cy3 and Cy5 according to the dual-color scheme. Our raw
data (no background correction) showed the following average
correlation coefficients among different pairs of 20 chips: 0.97 for
the Cy3 color channel, and 0.98 for the Cy5 color channel. In total,
97% of all pair-wise correlation coefficients were over 0.95.
Each probe was synthesized in
triplicate, so we estimated coefficients of variation for replicate
probes on the same array. In total, less than 3% of all genes had
coefficients of variation exceeding 20%.
Our arrays were hybridized
with the same material labeled in Cy3 and Cy5, so we calculated a false
positive discovery rate based on 2-fold change in these 20
same-versus-same comparisons. Among 247,920 ratio measurements, we
observed 551 ratios equal or exceeding 2-fold, thus, only 0.22% of
false positives.
Such high between- and
within-array reproducibility and low false positive rate have been
observed with raw data prior to background correction. Background
correction and simple ‘global’ normalization bring pair-wise
correlation coefficients between Cy3 and Cy5 color channels on the same
chip to average 0.992 (minimum 0.985) resulting in highly consistent
data (Fig. 3).
Figure 3. Scatter
plots of 15 replicate arrays hybridized with the same material
(Universal Human Reference RNA) labeled with Cy3 and Cy5. For each
array, X-axis is Log2(Cy3 Intensity). Y-axis is
Log2(Cy5 Intensity). Raw data has been
normalized and subjected to background correction.
WHAT SPECIAL EQUIPMENT
DO I NEED TO USE CUSTOMARRAY?
You will require a microarray scanner to image the chips after
processing. CustomArray is compatible with a number of imaging systems
that have been developed for spotted microarrays. We recommend a
slide-based, focusable scanner with a minimum of 5 um resolution. Axon,
Perkin Elmer, BioRad, and VIDAR Systems all have models compatible with our chips
(Axon GenePix 4000B, GenePix 4200A, GenePix 4200AL; Perkin Elmer
ScanArray® 4000 and 5000, ScanArray® Lite, ScanArray® Express; VIDAR Systems Revolution 4550). Contact
a technical support representative at CombiMatrix if you have questions
regarding the compatibility of your imaging system.
You may also need a rotisserie hybridization oven
to rotate arrays during hybridization. CombiMatrix produces a variety
of array holders for many common oven types used in the labs. All other
materials for processing should be readily available in a basic
molecular biology lab.
WHAT
MATERIALS ARE SHIPPED TOGETHER WITH CUSTOMARRAY 12K CHIPS?
Each CustomArray chip comes with the following items:
Hybridization chamber
Gasket (O-ring)
Two clamps
LifterSlip™ coverslip for imaging (Lifterslip™
is a trademark of Erie Scientific Corporation)
Imaging solution for array scanning.
All these material are intended for single use.
WHAT SORT OF MATERIAL DO I
NEED TO WORK WITH CUSTOMARRAY?
For each chip, you will need to prepare 5-10 micrograms of biotinylated
or fluorescently labeled nucleic acid (genomic DNA, cDNA, cRNA) using
your method of choice. Our technical support can provide some
guidelines depending on your type of array application.
HOW MANY PROBES/FEATURES
CAN I PUT ON A CUSTOMARRAY CHIP?
CustomArray 12K allows you to choose 12,000 distinct probes per array.
The arrays also have a set of 544 factory-built control probes for
quality control purposes.
ARE THERE ANY TOOLS FOR
PICKING LISTS OF GENES?
Genome sequencing data is available at The National Center for
Biotechnology Information (National Institute of Health) http://www.ncbi.nlm.nih.gov.
Tools that classify genes by functional categories are available at The
Cancer Genome Anatomy Project http://cgap.nci.nih.gov. This
system can be used to browse the Gene Ontology database and produce
lists of human and mouse genes. Importing lists of genes from
pre-existing data sets, or converting Unigene identifiers to accession
number formats can also be accomplished in batch using conversion tools
such as Matchminer http://discover.nci.nih.gov/matchminer.
WHAT
STEPS ARE INCLUDED IN THE COMBIMATRIX PROBE DESIGN ALGORITHM?
You can provide a file with GenBank accession numbers or nucleotide
sequences that correspond to genes of interest. Our probe design
software applies a set of proprietary algorithms that is based on the
generally accepted probe selection rules, and trained from experimental
data obtained with CustomArray chips. This software will perform
several subsequent iterations to select probes that are unique within
the defined gene set, are located in the specified part of genes, have
Tm and length within the specified range, do not form stable secondary
structure, and do not carry repeat sequences.
In addition, you can provide an ‘excluded targets’
file containing GenBank accession numbers or nucleotide sequences that
should not cross-hybridize with your probes. Typically, these sequences
correspond to highly abundant contaminants such as ribosomal RNA. It is
also possible to use this function to distinguish genes from a
homologous gene family, by inputting common parts of sequences as
excluded targets.
During the probe design process, the probe
specificity can be enhanced by screening the probe for uniqueness
within the known transcriptome. This additional screen is achieved
using a non-redundant database, Unigene, which contains a single entry
for each known gene of a particular genome. This screen is done
automatically after you specify the type of background organism from a
pull down menu (you may select up to three background organisms for one
run). In some cases, no relevant Unigene screen can be performed since
these databases are only available for a subset of organisms.
CAN
I CUSTOMIZE THE PROBE DESIGN ALGORITHM FOR ARRAY APPLICATIONS OTHER
THAN GENE EXPRESSION?
Our CustomArray chips can be used for a variety of different
applications that may require changes in the probe selection
algorithms. Our Design-on-Demand™ service can guide you through this
process and help to customize the probe design algorithm to your need.
Please see information on Design-on-Demand™
or email
for more information.
CAN I PUT MY OWN PROBES ON THE
ARRAY?
Yes. In addition to allowing the software to design probes for you, you
can also populate the chips with your own probes. For optimum
performance, you need to keep the length and Tm of the customer-defined
probes within the same range as for the rest of probes on the chip.
HOW LONG ARE THE
OLIGONUCLEOTIDE PROBES ON A CUSTOMARRAY?
Oligonucleotide probe length depends on application. For gene
expression studies we recommend to use 30 to 40-mer probes. SNP
analysis and re-sequencing may require short, 18 to 25-mer probes to
maximize mismatch sensitivity. The upper size limit for CustomArray
probes is set at 50-mer.
CAN I MIX GENOMES, SUCH AS
COMBINING PROBES TO VIRUS AND HOST GENES, ON ONE CHIP DESIGN?
Yes. Probes to multiple genomes can be used as long as the Tm of the
probes is within the same range (we recommend within five degrees for
optimal performance). You can specify up to three background
organisms/genomes for an automatic screening for probe uniqueness.
WHAT SORT OF CONTROLS DO I
NEED?
You can use spiking controls, negative controls, and positive controls,
such as housekeeping genes. Labeled spiking controls can be used to
evaluate array performance (between- and within-array variation), they
may be added to the labeled target samples prior to the hybridization
and should be in the 1-100 pM range. Negative controls should be
included for background calculation purposes. Housekeeping genes may be
used for normalization among arrays.
DOES
CUSTOMARRAY HAVE FACTORY-BUILT CONTROLS?
Yes, all CustomArray chips carry a set of factory-built controls
including two types of probes: negative controls (NC), and quality
controls (QC). In the factory layout, they are positioned as blocks in
different areas around the chip. A factory layout scheme with labeled
pre-built controls is shown in Figure 4. Several probe blocks are left
as ‘no-synthesis’ controls for QC purposes.
Both NC and QC probes could be used in your
experiments as negative controls for background calculation, providing
that they do not cross-hybridize with your material. The NC probes
labeled "plant," "bacteria," and "phage" are derived from Arabidopsis
thaliana, Agrobacterium tumifaciens, and phage lambda, respectively.
The exact sequences of these probes are shown in Table 1. You could
BLAST these control sequences against your genome of interest to
estimate cross-hybridization. If you work with plant and bacterial
genomes, you may need to design your own additional set of negative
controls instead of using the pre-built ones.
Figure
4. The factory layout of pre-built control probes (shown as red
squares).
Table
1: CustomArray 12K chip: factory-built control probes.
CAN
I REPLACE THE FACTORY-BUILT CONTROLS WITH MY OWN CONTROL PROBE SET?
No. These probes are essential for array quality control that is done
individually for each synthesized chip after manufacturing.
WHAT
SPIKING CONTROLS CAN I USE TO ESTIMATE CUSTOM ARRAY PERFORMANCE?
The factory-built controls from our standard factory layout can be used
as spiking probes, providing that they do not cross-hybridize with your
material. The sequences of these probes are shown in Table 1. You can
order oligonucleotides that are reverse-complimentary to the selected
control probes, and get them end-labeled (biotin or fluorescent dye,
should be the same labeling as in the planned experiments). Two or
three of these oligonucleotides would be sufficient as spiking controls
to add to the hybridization mixture.
Phage lambda spiking controls could be prepared
and used with the NC probes named ‘NC-SP-phage’ that are included in
the standard factory layout. Preparation of lambda spiking controls
from commercial phage lambda DNA includes several steps: PCR
amplification of lambda DNA fragments, cloning into a vector containing
the T7 RNA-polymerase promoter, sequencing of the cloned fragments, in
vitro transcription, and quality control of the resultant RNA.
You can also use commercially available Stratagene
spiking control mRNA samples made from 7 Arabidopsis
thaliana genes. Typically two or three spiking controls should be
sufficient. You need to select them and include the corresponding
probes from the following list into your array design files.
Please note that these controls would
cross-hybridize if you work with plant material.
WHAT
NEGATIVE CONTROLS CAN I USE?
Negative control probes are usually used to calculate background for
subtraction. The area surrounding spots and no synthesis spots cannot
be used for background calculation because they give an over-estimation
of background. You can use factory-built probes (both negative, NC, and
quality controls, QC) as negative controls, providing that they do not
cross-hybridize with your material. The NC probes labeled "plant,"
"bacteria," and "phage" are derived from Arabidopsis thaliana,
Agrobacterium tumifaciens, and phage lambda, respectively. If you work
with any of these organisms, you may need to provide your own set of
negative controls for probe design.
Plotting the distribution of NC spot intensities
can show how well particular NC probes perform in the context of you
array. For background subtraction, we suggest to use only NC probes
with low intensities (lowest 5-30%). If your arrays do not have an
appropriate set of negative controls with low signal intensities, it
might be better to omit the background subtraction step from data
analysis.
WHAT
POSITIVE CONTROLS CAN I USE?
You can provide a list of housekeeping genes for positive control probe
design. It should include genes with minimum variation in the
expression levels under the applied experimental conditions. Thus, it
is impossible to provide general recommendations applicable to all
systems. However, a list of potential candidates for human arrays can
be obtained here.
You need to select a relatively large set of
housekeeping genes to make the housekeeping gene approach less
sensitive to outlier data, because even such 'constantly' expressed
genes as actin still tend to have some variation in expression levels.
In addition, you need to keep this set constant for all arrays that you
plan to compare to each other. We recommend using at least 50 different
genes with sufficient numbers of replicates (at least three per gene).
Ideally, your set of housekeeping genes should contain both high and
low expressed ones to cover the full range of signal intensities
(except for the extremely high, close to saturation, and extremely low,
close to background levels).
CAN
I USE RIBOSOMAL RNA SEQUENCES AS POSITIVE CONTROLS?
We do not recommend it for gene expression studies in eukaryotic
organisms, because the standard mRNA amplification methods (based on
oligo(dT) primers) should dramatically reduce the proportion of rRNA in
the sample. This reduction may be non-uniform for different samples,
different amplification reactions, and different inputs of total RNA.
Variable results of this selection against rRNA make it a poor control
for array data normalization.
HOW
MUCH STARTING MATERIAL (RNA OR DNA) DO I NEED FOR AN ASSAY WITH
CUSTOMARRAY CHIPS?
CustomArray chips have been shown to work with a wide variety of
labeled target nucleic acids produced using different commercial kits.
So the amount of initial material that you need to start with depends
on the applied technique for target amplification and labeling. For
example, T7 polymerase-based RNA amplification kits (e.g. from Arcturus
and Ambion) result in very high degrees of amplification (especially
when used in double rounds). Thus you may start with as little as 100
ng of total RNA and still get enough material to hybridize with
CustomArray chips.
CAN
I USE BOTH SINGLE-COLOR AND DUAL-COLOR HYBRIDIZATION APPROACHES? HOW DO
I LABEL NUCLEIC ACID TARGETS FOR HYBRIDIZATION WITH CUSTOMARRAY CHIPS?
Yes, both approaches work for CustomArray chips. For the single-color
approach we recommend to label nucleic acid targets by biotin
incorporation, to hybridize arrays with the biotinylated targets, and
then to perform post-hybridization labeling using
streptavidin-conjugated fluorescent dye (e.g. Fluorolink Cy5-labeled
streptavidin from Amersham, Catalog # PA45001). For the dual-color
approach, two target nucleic acid samples may be directly labeled with
two different fluorescent dyes (e.g. Cy3 and Cy5) during amplification
reactions using commercially available kits (e.g. Ambion kits for cRNA
and cDNA labeling). For gene expression applications with eukaryotic
material we recommend to use the Ambion MessageAmp™ II aRNA
Amplification Kit (Catalog #1751).
Direct incorporation of fluorescent dye labels
into nucleic acids results in short shelf life of the labeled targets
which have to be utilized within one month from labeling. Only
biotinylated targets can be stored as long as unmodified nucleic acids.
The storage problem can be alleviated if you amplify and store nucleic
acid targets without labeling, and then apply a quick direct labeling
technology using Label IT® reagents from Mirus (http://www.mirusbio.com)
to obtain samples for array hybridization. This technology is based on
binding of a label to the guanine residues in DNA and RNA in a fast
non-enzymatic reaction. You can select a Label IT® Kit with a
fluorescent dye (e.g. Cy3 and Cy5) or biotin, and get your target
samples ready for hybridization in approximately 2 hours.
WHAT IS THE OPTIMUM TEMPERATURE
FOR HYBRIDIZATION?
Recommendations for hybridization buffers and temperatures are outlined
in our ‘Hybridization Protocol for CustomArray™12K
Chips & Preparation for Scanning', and they are
designed for gene expression analysis. Other applications may require
some optimization of these parameters. Please contact our technical
support specialists for further suggestions.
HOW MUCH NUCLEIC ACID DO I NEED
FOR THE HYBRIZATION?
You need 5 microgram biotinylated or fluorochrome-labeled nucleic acid
per array per color channel (10 microgram for the dual-color
experiments).
DO
I NEED TO ROTATE ARRAYS DURING HYBRIDIZATION?
Yes. Usage of a rotisserie oven or a rotating incubator is highly
recommended to ensure mixing during hybridization. You can introduce a
small air bubble into the hybridization chamber to improve the mixing
process during rotation. CustomArray chips can be attached to standard
rotisseries using various models of holders manufactured by
CombiMatrix. If you plan to buy a new rotisserie hybridization oven, we
recommend Fisher Scientific Isotemp Hybridization Incubators which can
be outfitted with a large CombiMatrix rotisserie holding 32 arrays.
CAN
I STRIP THE HYBRIDIZED TARGET NUCLEIC ACIDS AND RE-USE CUSTOMARRAY
CHIPS?
A single CustomArray™ microarray can be stripped three times using the
CombiMatrix CustomArray™ Stripping Kit. The CustomArray™ Stripping
protocol is based on chemical denaturation of DNA:DNA and DNA:RNA
hybrids between oligonucleotide probes on microarrays and corresponding
labeled targets. As a result of this denaturation, labeled targers are
removed from oligonucleotide probes and washed off microarrays. The
resultant stripped CustomArray™ microarrays can be used in the same way
as newly synthesized ones. In total, the CombiMatrix Stripping Kit
would enable you to use a single CustomArray™ microarray four times.
HOW DO I IMAGE MY ARRAY?
The CustomArray chip should be imaged “wet” to obtain the best
signal-to-noise ratio. A special imaging solution is provided with each
array shipment. Each array is also provided with a special LifterSlip™
coverslip that allows the array to be imaged under buffer. The
LifterSlip™ is used to protect the array from damage resulting from
direct contact with the glass, and to prevent excess liquid from
damaging the scanner. Please see our ‘Hybridization
Protocol for CustomArray™12K Chips & Preparation for Scanning’
for recommendations on how to prepare arrays for scanning. It is
critical to image the array in the correct orientation with the chip
facing the optics. It is not possible to image the array through the
slide.
CAN I USE MY OWN COVERSLIP
FOR IMAGING?
No. The LifterSlip™ must be used, or some damage may occur to the
array. The two raised edges, with white coating on the sides, prevent
the LifterSlip from directly touching the surface of the array. The
LifterSlip provides a 50 micron high cavity above the chip surface.
WHAT
SCANNERS DO I USE?
CombiMatrix recommends the following scanners:
Axon GenePix™ 4000B, 4200A, and 4200AL
AppliedPrecision: arrayWoRx®e Biochip Reader
PerkinElmer: ScanArray® 4000, ScanArray® 5000,
ScanArray® Lite, ScanArray® Express (please contact our technical
support about applicability of ScanArray Gx ®, ProScanArray®,
ProScanArray® HT)
VIDAR Systems: Revolution 4550
Other scanner models may also be used providing
that they have two essential features:
scan resolution of 5 micron
adjustable focus (manual, or the auto-focusing
feature)
WHAT AM I SEEING ON MY
SCANNED IMAGE?
Restrict the scan area to the array part (semiconductor chip) to
minimize scanning time. You will see a regular pattern of round circles
of varying intensity which correspond to the discrete features on the
array. Each spot is located within a dark square or rectangle bounded
by elements of the semiconductor chip. These elements show up as a
visible grid that does not normally interfere with data extraction. In
any case, the area outside of spots cannot be used for background
calculations (please see below for recommendations on background
calculation).
HOW
LARGE IS ONE SPOT ON CUSTOMARRAY?
Each spot is 44 micron in diameter, and the distance between centers of
two neighbor spots is 75 micron. Thus, one square of the semiconductor
grid that appears on CustomArray™ images is also 75 micron (Fig. 1).
HOW
MANY ROWS AND COLUMNS OF SPOTS AM I SEEING ON MY SCANNED IMAGE?
CustomArray™ 12K has 224 rows and 56 columns of spots.
CAN
I USE COMBIMATRIX SOFTWARE FOR DATA EXTRACTION?
Yes, we provide a free software application, Microarray Imager, for
data extraction; it can open most image formats. The detailed guide on
how to use it is located on our applications
website.
CAN
I USE THIRD-PARTY SOFTWARE FOR DATA EXTRACTION?
Yes, you can use data extraction software provided with your scanner.
In some cases, the appearance of a semiconductor grid around the spots
may cause problems with autotemplating. Our technical support team may
have solutions for some of these problems, or you can try our
Microarray Imager for data extraction.
HOW
DO I OBTAIN THE CHIP DESIGN FILE SHOWING POSITIONS OF PROBES ON MY
ARRAY?
A chip design file is a file in XML format that contains information
about your probes and their location on your CustomArrays. You can
access this file though our applications website by
registering and selecting "My Designs." Our data extraction software,
Microarray Imager, will directly accept the XML chip design file, so no
conversion is required. CombiMatrix also provides a tool for converting the chip design files
into GAL files. GenePix and some other programs for array
data extraction will accept the GAL format. CombiMatrix can assist
customers in converting to other file formats as well.
HOW DOES THE 12K CHIP DESIGN
FILE WORK WITH MICROARRAY IMAGER?
CombiMatrix stores probe location information in XML format in a file
called a chip design file. Microarray Imager understands this file and
can open most image formats, so no conversion of the input files is
required to extract data from images of our microarrays.
HOW DO I USE OTHER MEANS OF
IMAGING BESIDES THE MICROARRAY IMAGER?
CombiMatrix provides a tool for converting its chip design files into
GAL files. GenePix and some other programs for extracting data from
microarrays will accept this format. CombiMatrix can assist customers
in converting to other file formats as well.
HOW DO I EXTRACT DATA FROM MY
ARRAY?
The chip design file and the image of the hybridized chip from the
microarray scanner need to be opened in Microarray Imager (available on
the applications
website). It is critical to open the correct design file with
the correct chip image; the design file will have the annotations
associated for that specific chip. Microarray Imager will associate the
raw intensities for each spot on the chip with the appropriate
annotations once the data is extracted. Follow the instructions for
Microarray Imager available on this website.
HOW
DO I PERFORM ARRAY DATA NORMALIZATION?
Between-array normalization is performed to compensate for technical
variability. In most cases, a simple ‘global’ normalization technique
would be sufficient for CustomArrays. For each array, the median signal
intensity is calculated based on all probes. Then scale factors are
calculated for all arrays in the analyzed set to bring their median
signal intensities to the same level.
Several other normalization techniques may also be
used, such as the ones based on (1) a set of ‘housekeeping’ genes and
(2) probes with low intensity (the lowest 1-5% of signal intensities).
The housekeeping gene approach works if you are confident that the
selected genes are more or less constantly expressed in your material,
and if the gene set is large enough to compensate for outliers (at
least 100 of genes/probes). If the initial data normalization is done
based on a limited gene set, they need to be excluded from further data
analysis because the resultant normalized data for these genes could be
no longer considered statistically independent.
Background correction may be done after
normalization is completed.
HOW
DO I CALCULATE AND SUBTRACT ARRAY BACKGROUND?
Background calculation for CustomArray chips should be based on a set
of negative controls that do not hybridize with your material. Spots
with no synthesis, or the area outside of spots should not be used, as
these will give an over-estimation of background. You can provide your
own set of negative controls, or use the CustomArray set of
factory-built control probes. Our set includes probes named negative
controls (NC), and quality controls (QC), and both may be used,
providing that they do not cross-hybridize with your material. The NC
probes labeled "plant," "bacteria," and "phage" are derived from
Arabidopsis thaliana, Agrobacterium tumifaciens, and phage lambda,
respectively. If you work with plant and bacteria, you may see
significant cross-hybridization with our factory-built controls.
Plotting the distribution of NC and QC spot
intensities can show how well particular probes perform in the context
of your material. For background calculation, we suggest to use only
control probes with low intensities (lowest 5-30%), and then to
calculate their median signal intensity that can be used for
subtraction. All negative values are then converted into a common
default value (it can be set as 2x or 3x standard deviations of
background). This default should be the same for all arrays that you
plan to compare (not to create artificial ratios).
If your arrays do not have a special set of
negative controls with low signal intensities, you may use the lowest
5% of signals, or omit the background subtraction step from data
analysis.
WHAT SOFTWARE SHOULD I USE TO
INTERPRET CUSTOMARRAY RESULTS?
Data from Microarray Imager can be exported into a variety of third
party analysis software packages. Note that some formatting may be
required to upload data into third party software.
WHAT CAN I USE
CUSTOMARRAY FOR?
CustomArray can be used to order whole microarrays just like ordering
oligos! Initially, our platform has been tailored for gene expression
studies. At present, it is widely used for other applications,
including comparative genome hybridization (CGH), SNP analysis and
detection, re-sequencing, and chromatin immunoprecipitation (ChIP) on
chip. The web-based
system for designing chips can create optimized probes based
on a set of customer-defined sequences that are submitted as GenBank
accession numbers or raw sequence data. It can also accept
customer-specified oligonucleotide probe sequences. Our Design-on-Demand™ service
can help to change the probe design algorithm to tailor probes for your
particular application.
CAN I MAKE
DOUBLE-STRAND PROBES ON THE CUSTOMARRAY?
Yes. By incorporating a common linker at the 3' ends of all probes
(proximal to the array surface), one can anneal a primer that is
complementary to the linker and extend it with a DNA polymerase (such
as Klenow fragment of E. coli DNA polymerase I)
in buffer containing the dNTP mix.
CAN I MAKE ENZYMATIC ASSAYS
ON THE CHIP?
Yes, you can fill the hybridization chamber with an appropriate buffer
and perform most enzymatic reactions on the chip. You can phosphorylate
the 5’ ends of probes with T4 polynucleotide kinase, extend primers on
the probes with polymerases, ligate primers to probes with T4 DNA
ligase or E. coli DNA ligase, and perform restriction endonuclease
digestions, etc. This ability to perform enzymatic reactions on our
chips is the basis for development of new assays, such as Hyb &
Seq used for SNP analysis and re-sequencing. Please see Design-on-Demand™
for more information.
WHAT IF INTERNET EXPLORER
DOES NOT LAUNCH THE COMBIMATRIX SOFTWARE?
Java Web Start needs to be installed. If Java Web Start is not
installed, Internet Explorer will usually show a message that it cannot
open a file of type JNLP.
If Java Web Start has been installed and Internet
Explorer still will not launch CombiMatrix Software, it may be
necessary to clear the Internet Explorer temporary files.
WHAT OPERATING SYSTEMS DO I NEED TO
RUN COMBIMATRIX SOFTWARE?
CombiMatrix recommends Microsoft Windows XP or Microsoft Windows 2000.
WHAT IS THE SHELF LIFE OF A
CUSTOMARRAY?
Four months.
HOW SHOULD I STORE CUSTOMARRAY
CHIPS?
In a dark, dust free, dry environment at room temperature, or in a
dessicator.
CAN
I STORE HYBRIDIZED ARRAYS?
We do not recommend storage of hybridized arrays because it may cause a
decrease in signal and data quality. However, the hybridized arrays may
be stored for a short period of time (maximum a couple of days),
providing that they are kept wet, preferably with the hybridization
chambers filled with 2x PBS (or the imaging solution), well protected
from light (wrapped in foil), and at low temperature (4-8°C in
refrigerator).
If you have questions that have not been answered
here, please email:
Lifterslip™ is a trademark
of Erie Scientific Corporation.
ORDER 
