NetAffx Data Analysis Support—Troubleshooting

Registered users of the NetAffx™ Analysis Center can save their work for later use. After you have logged into the site, all of your query results are saved in the Sessions page. Subsequent logins allow you to select the desired queries and click the "View" button to retrieve the corresponding entries. Query results can also be saved to your desktop by selecting the Results page and then clicking the "Save" button.

The NetAffx™ Analysis Center supports the latest catalog arrays and a limited number of the discontinued arrays. Discontinued arrays are grouped under "Affymetrix Archival Target Sequences". In addition, the Analysis Center also provides the original UniGene build used for the probe set design. This build is grouped under "Archival UniGene".

For any particular array set, there are probe sets for only a subset of all possible sub-clusters. Moreover, some sequences are excluded because they are of low quality or are problematic (for example, too short or too repetitive), while other sequences have been added to the public databases since the array was designed. If you cannot find your sequence identifier, try running BLAST to search for a similar sequence or using the Probe Match tool. Please note: You have to search all the sub-clusters from a UniGene cluster before concluding that your sequence is missing in the sub-cluster databank.

Some array designs, such as the GeneChip™ Mu19K array, are based on the sequences in The Institute for Genomic Research (TIGR) databanks. Sub-cluster information for these probe arrays is considered TIGR's proprietary information and therefore cannot be distributed by us.

There can be a number of reasons for this. It is possible that the probe sequence has high homology with another unknown sequence, resulting in a high Mismatch-to-Perfect Match ratio. Another possibility is that the probe sequence on the array is correct for the majority of the cases, however, the sample may have a sequence variation that causes low specific binding to the Perfect Match, and high specific binding to the Mismatch. Regardless of the cause, the redundancy of the number of probes representing a sequence on GeneChip™ expression arrays negates any significant impact this may have on the final interpretation of the data.

A high background implies that impurities, such as cell debris and salts, are binding to the probe array in a nonspecific manner and that these substances are fluorescing at 570 nm (the scanning wavelength). This nonspecific binding causes a low signal to noise ratio (SNR), meaning that genes for transcripts present at very low levels in the sample may incorrectly be called Absent. High background creates an overall loss of sensitivity in the experiment.

Files in CD-ROM format are copied to the hard drive in read-only mode. MAS requires that this attribute be removed. To do this, open NT Explorer and select the file(s) you copied from the CD. Click the right mouse button and select Properties. Clear the Read-only check box near the bottom of the Properties screen and click OK.

There could be a number of reasons for this. It is possible that this probe sequence has high homology with another unknown sequence resulting in a high mismatch-to-perfect match ratio. Another possibility is a mutation or set of mutations in the sequence of the target transcript which causes specific binding to the Mismatch. Regardless of the cause, the built-in redundancy using multiple probe pairs to represent a single sequence on the probe array mitigates any significant impact on the final interpretation of the data.

No, only the Perfect Match sequences are provided through the Analysis Center. However, there is a simple relationship between the Perfect Match and the Mismatch probes: the Mismatch probes are identical to the Perfect Match probes, except for a simple substitution of the complementary base (homomeric transversion) at the 13th base position of the Mismatch sequence.

The following example illustrates the relationship between the Perfect and Mismatch probe sequences:

Discard the results if the number of matching probes constitutes less than 70% of the total number of probes in a probe set. You can find the total number of probes matching the query sequence in the "# Probes Matching Query" column of the Probe Match output.

As an example, let the topic be "cardiovascular disease" and the species of interest is "human".

• Simply type into the Query text field: "cardiovascular".
• Be sure that Human is selected in the Species tree.
• Click “Submit!”

The Array Comparison report matrix shows the number of genes associated with Cardiovascular covered by each array (in the diagonal). The matrix also shows the number of Cardiovascular-associated genes that are covered by a pair of arrays (off-diagonal).

As an example, let the topic be "apoptosis" in "mouse". We are interested in its translation to "human".

• Simply type into the Query text field: "apoptosis".
• Be sure that Human and Mouse are selected in the Species tree.
• Click “Submit!”

The Array Comparison report matrix shows the number of genes associated with Apoptosis covered by each array (in the diagonal). The matrix also shows the number of Apoptosis-associated genes that are covered by a pair of arrays (off-diagonal). Homologene is used when comparing two different species.

If a catalog array of interest cannot be found, try de-selecting "Only Most Recent Arrays". Also, be sure that the parent term in the Species tree is expanded.