Excess significance bias in repetitive transcranial magnetic stimulation literature for neuropsychiatric disorders, bioRxiv, 2019-04-23

ABSTRACTIntroductionRepetitive transcranial magnetic stimulation (rTMS) has been widely tested and promoted for use in multiple neuropsychiatric conditions, but as for many other medical devices, some gaps may exist in the literature and the evidence base for rTMS clinical efficacy remains under debate. We aimed to empirically test for an excess number of statistically significant results in the literature on rTMS therapeutic efficacy across a wide range of meta-analyses and to characterize the power of studies included in these meta-analyses.MethodsBased on power calculations, we computed the expected number of “positive” datasets for a medium effect-size (standardized mean difference, SMD=0.30) and compared it with the number of observed “positive” datasets. Sensitivity analyses considered small (SMD=0.20), modest (SMD=0.50), and large (SMD=0.80) effect sizes.Results14 meta-analyses with 228 datasets (110 for neurological disorders and 118 for psychiatric disorders) were assessed. For SMD=0.3, the number of observed “positive” studies (n=94) was larger than expected (n=35). We found evidence for an excess of significant findings overall (p<0.0001) and in 814 meta-analyses. Evidence for an excess of significant findings was also observed for SMD=0.5 for neurological disorders. 0 (0 %), 0 (0 %), 3 (1 %), and 53 (23 %) of the 228 datasets had power >0.80, respectively for SMDs of 0.30, 0.20, 0.50, and 0.80.ConclusionMost studies in the rTMS literature are underpowered. This results in fragmentation and waste of research efforts. The somewhat high frequency of “positive” results seems spurious and may reflect bias.Trial Registration PROSPERO 2017 CRD42017056694

biorxiv scientific-communication-and-education 0-100-users 2019

pathwayPCA an R package for integrative pathway analysis with modern PCA methodology and gene selection, bioRxiv, 2019-04-23

ABSTRACTWith the advance in high-throughput technology for molecular assays, multi-omics datasets have become increasingly available. However, most currently available pathway analysis software provide little or no functionalities for analyzing multiple types of -omics data simultaneously. In addition, most tools do not provide sample-specific estimates of pathway activities, which are important for precision medicine. To address these challenges, we present pathwayPCA, a unique R package for integrative pathway analysis that utilizes modern statistical methodology including supervised PCA and adaptive elastic-net PCA for principal component analysis. pathwayPCA can analyze continuous, binary, and survival outcomes in studies with multiple covariate andor interaction effects. We provide three case studies to illustrate pathway analysis with gene selection, integrative analysis of multi-omics datasets to identify driver genes, estimating and visualizing sample-specific pathway activities in ovarian cancer, and identifying sex-specific pathway effects in kidney cancer. pathwayPCA is an open source R package, freely available to the research community. We expect pathwayPCA to be a useful tool for empowering the wide scientific community on the analyses and interpretation of the wealth of multiomics data recently made available by TCGA, CPTAC and other large consortiums.

biorxiv bioinformatics 0-100-users 2019

 

Created with the audiences framework by Jedidiah Carlson

Powered by Hugo