Since these SNPs are determined empirically from ALCHEMY or the vendor's software, we required that a SNP fail one or both of these criteria for both ALCHEMY and the vendor's software (Beadstudio or BRLMM-P) in order to be excluded.
To detect a difference of at least 20% between the groups with a two-sided test of proportions, alpha of 0.017 (three comparisons 0.05/3), and 80% power we determine using a statistical software that we require at least 124 patients in each group.
Because some of the software we compared required reference data, we provided samples of HapMap Phase II release 22 as the reference samples; those samples were, thus, excluded in our testing set.
To detect a difference of at least 30% between the groups with a two-sided test of proportions, alpha of 0.017, and 80% power we determine using statistical software [ 45] that we require at least 52 patients in each group.
Given our need to image an array over the course of an experiment, we require software that can apply the same grid to an entire set (or stack) of images.
We should check if we require the software in a longer run, if not, we should go for getting it on lease and save some money out of it.
The software we've selected requires registration or payment.
Formal approaches to software development require that we correctly describe (or specify) systems in order to prove properties about our proposed solution prior to building it.
It reduces the cost of a transaction many fold, is low cost to implement, uses open source software we've developed — but it requires cooperation amongst the participants that are traditionally adversarial.
The use of certain software may require changes.
Software may require more resources than the system can provide.
