Background: The detection of true significant cases under multiple testing is becoming a fundamental issue when analyzing high-dimensional biological data. Unfortunately, known multitest adjustments reduce their statistical power as the number of tests increase. We propose a new multitest adjustment, based on a sequential goodness of fit metatest (SGoF), which increases its statistical power with the number of tests. The method is compared with Bonferroni and FDRbased alternatives by simulating a multitest context via two different kinds of tests: 1) one-sample t-test, and 2) homogeneity G-test.

The scale-of-choice effect and how estimates of assortative mating in the wild can be biased due to heterogeneous samples

Two sympatric snail ecotypes (RB and SU) of Littorina saxatilis from exposed rocky shores of NW Spain differ in many life history traits, but classical morphometric analysis has failed to reveal significant shell shape differences between them. We used geometric morphometric methods on landmark data from digitized shell images to study size and shape components in both ecotypes at two localities. The results showed significant differences between ecotypes in both shell size and shape (both uniform and non-uniform components). Allometry was also detected for some component of the local variation in shape, although it did not explain the observed differences between ecotypes. The SU ecotype had a relatively rounded shell shape with a big aperture, whereas the RB ecotype had higher spire and smaller aperture. We suggest that shape differentiation is correlated with adaptive differences between ecotypes. 1997; Cruz et al., 2001). Johannesson et al. (1993) used 13 shell measurements and principal component analysis to study shell form in several populations of these two ecotypes. These authors observed significant differences in the first principal component (argued to be mainly due to size variation because all loadings were positive and of similar magnitude), but they could not detect any significant difference in other principal

In quantitative proteomics work, the differences in expression of many separate proteins are routinely examined to test for significant differences between treatments. This leads to the multiple hypothesis testing problem: when many separate tests are performed many will be significant by chance and be false positive results. Statistical methods such as the false discovery rate method that deal with this problem have been disseminated for more than one decade. However a survey of proteomics journals shows that such tests are not widely implemented in one commonly used technique, quantitative proteomics using two-dimensional electrophoresis. We outline a selection of multiple hypothesis testing methods, including some that are well known and some lesser known, and present a simple With the advent of high throughput genomics approaches, researchers need appropriate bioinformatic and statistical tools to deal with the large amounts of data generated. In quantitative proteomics work, differences in expression of many individual proteins between treatments or samples might need to be tested. Researchers must then address what has come to be known as the multiple hypothesis testing problem. Suppose 500 features such as protein spots in a two-dimensional electrophoresis (2-DE) 1 experiment, or mass spectrum features relating to protein or peptide abundance, are each compared between treatments using a t test. If the conventional a priori significance level of ␣ ϭ 0.05 is used, then 5% or about 25 significant features are expected to occur just by chance even if the null hypothesis of no treatment effect is true for all 500 features. Thus it is easier to make a false positive error when picking out significant results in an experiment with multiple features, than when considering one feature in isolation.A variety of statistical methods have been devised to deal with the multiple hypothesis testing problem. These are applicable in quantitative proteomics. In this paper we use examples from 2-DE proteomics to demonstrate these methods. In this technique, the intensity of signal from protein spots on 2-DE gels is measured and compared between gels. Use of the word "spot" is obviously not synonymous with use of the word "protein" in that it does not encompass all forms of a given protein such as alternatively spliced variants and posttranslational modification variants that might form spots in different positions on the gel. The multiple testing approach is introduced with the following example. Table I shows simulated data for a model of a 2-DE proteomics experiment in which 500 spots have been compared between two treatments using the t test. The third column gives p values significant at ␣ ϭ 0.05 sorted from low to high. A threshold line is shown drawn under spot 70. This has been selected arbitrarily for illustration of some properties of a threshold. The p values for the spots above the threshold are all less than ␣ ϭ 0.05 but we cannot declare them to be significant at the ␣ ϭ 0.05 level because of the multiple hypot...

Populations of the marine gastropod Littorina saxatilis from exposed rocky shores of NW Spain provide one of the few putative cases of sympatric ecological speciation. Two ecotypes with large differences in shell morphology and strong assortative mating are living at different vertical levels of the shore separated by a few meters. It has been hypothesized that shell size is the main determinant for the reproductive isolation observed between the ecotypes, and that several shell shape traits are subject to divergent natural selection and are responsible for the adaptation of each ecotype to its respective habitat. Using embryos extracted from wild females we obtain estimates of genetic variation for shell size and shape and compare them with those from neutral molecular markers. Estimates of heritability are significantly larger for the ecotype found in the upper shore than for that in the lower shore, in concordance with a similar result observed for heterozygosity of neutral markers. The large genetic differentiation between ecotypes for the shell traits, contrasting the smaller close to neutral differentiation between populations of the same ecotype, supports the implication of the traits in adaptation.