Naomi > Currently I am telling the biologists to consider microarrays as > screening experiments. Mostly, they use the results for second stage > analyses, which may be: ok. That's your choice. > e.g. statistical analyses such as clustering etc > bioinformatics analyses such as GO, BLAST or sequence analyses > lab analyses such as Northern blots, in situs, etc > > Given the huge number of genes on most arrays, I do want a reasonably > reliable method of screening. On the other hand, I sometimes just rank > the genes by test score, rather than attempt to determine some > suitable alpha-level, FDR or FNR. Fine. MTA is a fine statistical research area. Good for stats departments the world over. Solid, employable kind of work. > Incidentally, distinguishing between technical replicates and > biological replicates can make a huge different to ANOVA test scores, > so I think we should insist that our analyses should account for this. I agree with this observation. Certainly there's a difference between biological and technical replication, and I've seen instances where the v(B)>V(T) and where v(T)>v(B). The point is: biologists initially WANTED to use arrays as huge northerns -- as huge screening experiments. Enabling them to do this makes us...enablers (aka pushers). We're NOT facing the future when we serve up single gene results. I've got at least 1 fine argument that says ordering genes by T-stat (or any other univariate score based on single gene array results) gives a MEGA CRAPPY ordering of 'improtante' for the genes. So. The revolution, she is coming. Gonna wipe out molec biology as we know it. No more single gene science. Don't be first against the wall when the revolution comes. Vive la revolcion CASHAW > > --Naomi > > > At 09:09 AM 12/18/2003, Stephen Henderson wrote: > >> I agree with some of WHAT you say CHAD, the PROBLEM is THAT MOST >> multiVARIATE methods are BUILt on top OF the marginal tests. FOR >> instance >> machine learning methods are based on gene subsets for each of k CROSS >> validations. USE of the appropriate TEST (fold/T/F/cyber-T/etc..)for >> subset >> selection is IMHO the most IMPORTANT!! choice . >> >> >> Stephen >> >> >> ********************************************************************** >> This email and any files transmitted with it are >> confidentia...{{dropped}} >> >> _______________________________________________ >> Bioconductor mailing list >> Bioconductor@stat.math.ethz.ch >> https://www.stat.math.ethz.ch/mailman/listinfo/bioconductor > > > Naomi S. Altman 814-865-3791 (voice) > Associate Professor > Bioinformatics Consulting Center > Dept. of Statistics 814-863-7114 (fax) > Penn State University 814-865-1348 (Statistics) > University Park, PA 16802-2111 > > > >