 |
|
|
Array results can potentially be compared directly. However, it is important to check the following important elements before doing so:
1. Experimental design strategy should be the same at various sites.
2. Identical target labeling protocols should be followed, and yields from cDNA and IVT reactions should be within the same range as specified for that study.
3. Same algorithm parameters are used.
4. Similar results from 3'/5' ratios, background, noise, and scaling factors. Check arrays for scratches and even hybridization/staining.
5. Comparability of results obtained from different operators should be evaluated before including their results in the same study. |