Motivation: Survival prediction from gene expression data and other high-dimensional genomic data has been subject to much research during the last years. These kinds of data are associated with the methodological problem of having many more gene expression values than individuals. In addition, the responses are censored survival times. Most of the proposed methods handle this by using Cox's proportional hazards model and obtain parameter estimates by some dimension reduction or parameter shrinkage estimation technique. Using three well-known microarray gene expression data sets, we compare the prediction performance of seven such methods: univariate selection, forward stepwise selection, principal components regression (PCR), supervised principal components regression, partial least squares regression (PLS), ridge regression and the lasso. Results: Statistical learning from subsets should be repeated several times in order to get a fair comparison between methods. Methods using coefficient shrinkage or linear combinations of the gene expression values have much better performance than the simple variable selection methods. For our data sets, ridge regression has the overall best performance. Availability: Matlab and R code for the prediction methods are available at http://www.med.uio.no/imb/stat/bmms/software/microsurv/. Contact:hegembo@math.uio.no

Fisheries catches worldwide have shown no increase over the last two decades, while aquaculture has been booming. To cover the demand for fish in the growing human population, continued high growth rates in aquaculture are needed. A potential constraint to such growth is infectious diseases, as disease transmission rates are expected to increase with increasing densities of farmed fish. Using an extensive dataset from all farms growing salmonids along the Norwegian coast, we document that densities of farmed salmonids surrounding individual farms have a strong effect on farm levels of parasitic sea lice and efforts to control sea lice infections. Furthermore, increased intervention efforts have been unsuccessful in controlling elevated infection levels in high salmonid density areas in 2009–2010. Our results emphasize host density effects of farmed salmonids on the population dynamics of sea lice and suggest that parasitic sea lice represent a potent negative feedback mechanism that may limit sustainable spatial densities of farmed salmonids.