Assessing the validity of blood-based gene expression profiles for the classification of schizophrenia and bipolar disorder: A preliminary report Ming T. Tsuang, Nadine Nossova, Tom Yager, Min-Min Tsuang, Shi-Chin Guo, Kou Ge Shyu, Stephen J. Glatt, C.C. Liew American Journal of Medical Genetics Part B: Neuropsychiatric Genetics Volume 133B, Issue 1 , Pages1 - 5 Please, read the manuscript and the methods… And the “power of discrimination” ….“Outlier samples that may confound the reference dataset were identified using neural network analysis (EasyNN-Plus version 4.0 g, Neural Planner Software, Inc.) and removed prior to performing the logistic regression computations.” Needless to say, the data has been never replicated my anyone (although many have tried it). http://www3.interscience.wiley.com/cgi-bin/fulltext/109865056/HTMLSTART Karoly Mirnics, MD I was embarrassed by a recent publication in the Journal of the American Society of Nephrology....of which Eric Nielson is the editor-in-chief. In this post hoc analysis, they reported a significant treatment effect of rosuvastatin in an ex post facto subgroup on one component of the primary composite endpoint (http://www.ncbi.nlm.nih.gov/pubmed/21566054). The parent trial detected no effect on the primary composite overall. I had never wished more that JASN published letters to the editor. Hope all is well, Jim _____________________________________________________________ James P. Smith, MD, MS Medical Officer, Division of Metabolism & Endocrinology Products Food and Drug Administration / CDER WO-22, Rm 3361 10903 New Hampshire Ave. Silver Spring, MD 20993-0002 (301) 796-1317 (office) (301) 796-9712 (fax) james.p.smith@fda.hhs.gov Video of Ransohoff 2007 presentation: http://videocast.nih.gov/summary.asp?Live=6157&bhcp=1 ----------------------- http://www.buzzfeed.com/bengoldacre/deworming-trials Crowd-sourcing analysis and good example of a different result with re-analysis:http://www.nature.com/news/crowdsourced-research-many-hands-make-tight-work-1.18508 http://www.economist.com/news/science-and-technology/21672076-honest-disagreement-about-methods-may-explain-irreproducible-results-other?frsc=dg%7Ca ----------------------- http://www.nature.com/news/reproducibility-a-tragedy-of-errors-1.19264http://www.nature.com/news/reproducibility-a-tragedy-of-errors-1.19264 http://www.nature.com/news/blind-analysis-hide-results-to-seek-the-truth-1.18510http://www.nature.com/news/blind-analysis-hide-results-to-seek-the-truth-1.18510 --------------------------------------------------- Claiming effectiveness of call patients after discharge when they were really comparing those contacted to those not able to be contacted: http://onlinelibrary.wiley.com/enhanced/doi/10.1002/phar.1630 http://www.nature.com/news/1-500-scientists-lift-the-lid-on-reproducibility-1.19970 --------------------------------------------------- Important editorial: http://reason.com/archives/2016/08/26/most-scientific-results-are-wrong-or-use Covers what goes wrong when investigators are left to figure out what to research --------------------------------------------------- http://www.nature.com/mp/journal/vaop/ncurrent/full/mp201744a.html Collaborative meta-analysis finds no evidence of a strong interaction between stress and 5-HTTLPR genotype contributing to the development of depression They also found that the main effect of the variant was zilch and that stressors were strong. --------------------------------------------------- Incentives smi17nat.pdf + others --------------------------------- https://geneticliteracyproject.org/2020/02/25/remember-the-questionable-study-claiming-glyphosate-boosts-cancer-risk-41-lead-author-reasserts-her-claim-epa-debunks-it-and-we-take-a-second-look/