@iskander Thanks, I hadn’t grasp that major subtlety. Re normalization, I guess you also need to avoid normalization on the full data? You would run any preprocessing like scaling on train/test separately to avoid similar problems?
@mikelove Late to the party but I also have some codes to share for the blue place, happy to connect with interested people by DM. There is also an invite pool here from @micahgallen
https://t.co/KgU0RiLKZL
@mikelove Late to the party but I also have some codes to share for the blue place, happy to connect with interested people by DM. There is also an invite pool here from @micahgallen
https://t.co/KgU0RiLKZL
I am thrilled to announce the release of my first R/Bioconductor package, "easylift."
This library simplifies the genomic liftover process, interfaces seamlessly with GRanges and UCSC chain files, and integrates with BiocFileCache.
https://t.co/YWEb6mmNYO
It amazes me that "Statistical analysis" are single subsection of "Methods" appendices in #microbiome papers. Y'all have separate sections for "Sample storage" (with careful justification), "DNA extraction" (w/ careful justification), "qPCR" (w/ careful justification)... 1/6
Anyone else want invite codes for MLSky / ImmunoSky / MedSky / GeneSky?
(genomics & immunology pretty active over there, could use some cliques of machine learning / deep learning folks to busy up that feed)
@statsepi Publication pressure ‘trickles down’ to academics, residents, med. students, HS . High pressure, hi volume/hi speed ‘academic’ environs grooming bad methods habits, ‘research’ becomes devalued & hollow word. How do we break the vicious circle of hype, noise and research waste?
@statsepi I was once in a department seminar where a colleague presented results from a clinical cohort, and went over sample swapping errors that we had to correct. After the talk a professor came to him and said to not do that again because that make us look sloppy
every time I look for an illustration of my work in #Bioinformatics#rstats, I stumble on the mighty *laptop with double helix on screen*. Do people usually visualize all their NGS data as a double helix DNA, am I doing something wrong?
#crowdresearch is awesome! Do you love #tidyverse and #omic' research?
There are 20+ open challenges still waiting for a champion. Join our effort, solve one (or many), and become part of our upcoming #publication!
https://t.co/hBjpoysOG4
#tidyomics@Bioconductor