What started as a 1 day side project turned into a nearly 3 year long lab analysis. "Microbiome differential abundance methods produce disturbingly different results across 38 datasets" https://t.co/56rVFk97fN
bioBakery 3 is out in @eLife! It describes our latest versions of MetaPhlAn 3, HUMAnN 3, StrainPhlAn 3, PanPhlAn 3 and more from @cibiocm and @hutlab ! Huge thanks to so many but especially to Francesco Beghini @chocophlan and Lauren McIver! https://t.co/rPMfMRWLzT
#Statistics thought of the day: if you are not certain how to transform a continuous dependent variable to make assumptions work (e.g. normality of residuals) use a semiparametric method that is transformation-invariant, e.g. proportional odds/hazard model. #rmscourse#bbrcourse
New R package rmsb: Bayesian counterpart to the rms (regression modeling strategies) package now on CRAN. Uses pre-compiled @mcmc_stan code. Information including lots of examples at https://t.co/XSMsQyqA1J #rmscourse @vandy_biostat @VUDataScience
In addition, you can now also set discrete colour/fill scale defaults via a vector (or list) of colour codes. Super useful for providing colour-blind safe defaults! #rstats
Woohoo -- I finally carved out some time today to make a Wiki to answer the most common questions I get about my #R#rstats package `breakaway` 🥳😻🎀
Check it out here:
https://t.co/upLmi0h3CS
#rstats#microbiome#microbialdiversity
HUMAnN 3.0 (alpha) is out! Functionally profile #metagenomes or #metatranscriptomes at species-level resolution. Designed in tandem with MetaPhlAn 3.0, more options for fine-tuning tiered search, 2x the species #pangenomes and 3x the gene families. Enjoy!
https://t.co/EU6ZsDI9LM
If you like using propensity matching, think again. See
https://t.co/YVWbwr1d7K or https://t.co/aAGdRvKZIK by @PhilDawid or https://t.co/rsz27yAJiH by @kinggary
Got observational data with time-varying treatments and confounders?
Want to estimate the effects of sustained treatment
strategies?
Then you'll enjoy this gentle introduction to the gfoRmula package in #rstats.
Led by Sean McGrath & Victoria Lin
https://t.co/Q9lVjCbWoI
@f2harrell @vandy_biostat Hi Frank @f2harrell ! Thanks for sharing the blog post. I hope people find it helpful. Look forward to digging into some of the new Bayesian functionality soon!