블록디자인에 의한 cDNA Microarray 실험에 관한 연구

Abstract

We need more efficient designs when block design is not efficient. In general , Latin square and Youden square designs many times are not suitable because they require more replications which is expensive. If we do not need to estimate some interactions then we can confound them and lose all information on those interactions (many times high order interactions are zero so we don't need to estimate them). By losing information on some interactions, we gain on the information (or better precision) on main effects, two-factor interactions etc. So, single replication is good for economy and confounded row-column design is good for improving precision. Also, balanced factorial designs are introduced for cDNA microarray experiments. Single replicate designs obtained using the classical method of confounding are shown to be particularly useful for deriving suitable balanced designs for cDNA microarrays. Classical factorial designs obtained using methods other than the method of confounding are also shown to be useful. The paper provides a systematic method of deriving designs for microarray experiments as opposed to algorithmic and ad-hoc methods, and generalizes several of the microarray designs given recently in literature. Also designs for factorial cDNA microarrays realized by programing.