MOLECULAR BIOLOGY & GENETICS On the low reproducibility of cancer studies