Searching for robust associations with a multi-environment knockoff filter.

This paper develops a method based on model-X knockoffs to find conditional associations that are consistent across environments, controlling the false discovery rate. The motivation for this problem is that large data sets may contain numerous associations that are statistically significant and yet misleading, as they are induced by confounders or sampling imperfections. However, associations replicated under different conditions may be more interesting. In fact, consistency sometimes provably leads to valid causal inferences even if conditional associations do not. While the proposed method is widely applicable, this paper highlights its relevance to genome-wide association studies, in which robustness across populations with diverse ancestries mitigates confounding due to unmeasured variants. The effectiveness of this approach is demonstrated by simulations and applications to the UK Biobank data.

Gene hunting with hidden Markov model knockoffs.

Modern scientific studies often require the identification of a subset of explanatory variables. Several statistical methods have been developed to automate this task, and the framework of knockoffs has been proposed as a general solution for variable selection under rigorous Type I error control, without relying on strong modelling assumptions. In this paper, we extend the methodology of knockoffs to problems where the distribution of the covariates can be described by a hidden Markov model. We develop an exact and efficient algorithm to sample knockoff variables in this setting and then argue that, combined with the existing selective framework, this provides a natural and powerful tool for inference in genome-wide association studies with guaranteed false discovery rate control. We apply our method to datasets on Crohn's disease and some continuous phenotypes.

HPV self-sampling in CIN2+ detection: sensitivity and specificity of different RLU cut-off of HC2 in specimens from 786 women.

AIMS: Mortality for cervical cancer varies between the different regions of the world, with high rates in low-income countries where screening programmes are not present and organised. However, increasing screening coverage is still a priority in all countries: one way to do that is to base screening on self-sampled screening. The success of a self-sampling screening strategy depends on capacity to recruit unscreened women, on the performance and acceptability of the device and on the clinical performance of the high-risk human papillomavirus (HPV) test. METHODS: This study based on 786 enrolled women investigates the best cut-off value of Hybrid Capture 2 HPV test (HC2) for self-sampled specimens in terms of sensitivity and specificity. RESULTS: In this population, we found that the sensitivity and the specificity for cervical intraepithelial neoplasia grade 2 or more detection of HC2 performed on self-sampled specimens were 82.5% and 82.8%, respectively considering the relative light units (RLU) cut-off value of 1. Increasing the cut-off value the sensitivity decreases and the specificity raises and the best area under the curve for the RLU cut-off value is 1. CONCLUSIONS: Our results confirm that the cut-off value of 1 suggested by Qiagen for PreservCyt specimen is the best cut-off value also for self-sampled specimens.