|
*[[Permutations]] are calculated based on the number of samples
*Block Permutations
**Blocks are batches of microarrays; for example for eight samples split into two groups (control and affected) there are 4!=24 permutations for each block and the total number of permutations is (24)(24)= 576. A minimum of 1000 permutations are recommended;<ref name="R1"/><ref name="R2">{{cite journal | last1 = Dinu, | first1 = I. P., | last2 = JD; | first2 = | last3 = Mueller, | first3 = T; | last4 = Liu, | first4 = Q; | last5 = Adewale, | first5 = AJ; | last6 = Jhangri, | first6 = GS; | last7 = Einecke, | first7 = G; | last8 = Famulski, | first8 = KS; | last9 = Halloran, | first9 = P; | last10 = Yasui, | first10 = Y. (| year = 2007). "| title = Improving gene set analysis of microarray data by SAM-GS." | url = | journal = BMC Bioinformatics | volume = 8: | issue = | page = 242. }}</ref><ref name="R3">{{cite journal | last1 = Jeffery, | first1 = I. H., | last2 = DG; | first2 = | last3 = Culhane, | first3 = AC. (| year = 2006). "| title = Comparison and evaluation of methods for generating differentially expressed gene lists from microarray data." | url = | journal = BMC Bioinformatics | volume = 7: | issue = | page = 359. }}</ref>
the number of permutations is set by the user when imputing correct values for the data set to run SAM
*Correlates expression data to clinical parameters<ref name="R6"/>
*Correlates expression data with time<ref name="R1"/>
*Uses data permutation to estimates False Discovery Rate for multiple testing<ref name="R7"/><ref name="R8"/><ref name="R6"/><ref name="R5">{{cite journal | last1 = Larsson, | first1 = O. W., C; | last2 = Timmons, | first2 = JA. (| year = 2005). "| title = Considerations when using the significance analysis of microarrays (SAM) algorithm." | url = | journal = BMC Bioinformatics | volume = 6: | issue = | page = 129. }}</ref>
*Reports local false discovery rate (the FDR for genes having a similar d<sub>i</sub> as that gene)<ref name="R1"/> and miss rates <ref name="R1"/><ref name="R7"/>
*Can work with blocked design for when treatments are applied within different batches of arrays<ref name="R1"/>
==References==
{{reflist|refs=
<ref name="R6">{{cite journal | last1 = Tusher, | first1 = V. G., | last2 = Tibshirani | first2 = R. Tibshirani,| display-authors = 2 | last3 = et al. (| year = 2001). "| title = Significance analysis of microarrays applied to the ionizing radiation response." Proceedings| ofurl the National Academy= of Sciences 98(9): 5116–5121. [http://www-stat.stanford.edu/~tibs/SAM/pnassam.pdf] | format = PDF | journal = Proceedings of the National Academy of Sciences | volume = 98 | issue = 9| pages = 5116–5121 }}</ref>
<ref name="R7">{{cite journal | last1 = Zang, | first1 = S., | last2 = Guo | first2 = R. Guo,| display-authors = 2 | last3 = et al. (| year = 2007). "| title = Integration of statistical inference methods and a novel control measure to improve sensitivity and specificity of data analysis in expression profiling studies." | url = | journal = Journal of Biomedical Informatics | volume = 40( | issue = 5):| pages = 552–560 552–560}}</ref>
}}
| 