Somnath, I'm not familiar with this error, but will look into this. As for computing a pvalue for each sample/geneset, bootstrapping is the method we offer. With bootstrapping turned on, it will return a list of size K where each k is the bootstrapped enrichment matrix. You can then use this to compute your statistic. I'll provide more detail once I track down this error. -justin -- Justin Guinney, Ph.D. Research Scientist Sage Bionetworks Fred Hutchinson Cancer Research Center 1100 Fairview Ave North Seattle, WA 98109 206 667-2146 (b) 919 357-4762 (c)