CoDaWork2019

📌 Convocatòria d'ajudes econòmiques de la @SEIO_ES per assistir al congrés CoDaCourse de juliol del 2020 a l'@univgirona cc/ @socestadistica. Més informació sobre el congrés 👉 compositionaldata.com/codacourses.php

I am really excited to be defending my PhD dissertation, on decision analysis under uncertainty for #sustainabledevelopment, this Thursday 31st October at 9.30 at the @la_UPC! If you are in Barcelona, please feel free to join! 😃

"Education isn't something you can finish." -- Isaac Asimov (1920 - 1992)

The medical student shadowing me looks over my clinic schedule. She sighs exaggeratedly. “Doctor T, your YOUNGEST patient today is 70. Don’t you get tired of only seeing older patients?” I look down at my patient list, and smile. I don’t see what she sees... 1/

Update: our effort to compile crowdsourced list of references on some "Common Statistical Myths" is now a wiki on DataMethods - please feel free to go there and add your own topics and/or references! discourse.datamethods.org/t/reference-co…

TIL ⏩ if you knit an #rstats #rmarkdown to html, and use the DT package (interface to the DataTables JavaScript library), you can add buttons for readers to download the data as csv + excel files 🖱️🗄️ rstudio.github.io/DT/003-tableto… (h/t to @abtran 🎩 andrewbtran.github.io/NICAR/2018/wor…)

Call to #codawork2019 participants and #CoDA community with bioinformatic applications: @gbgloor, @tpq__ and I are organizing a special issue of NARGAB academic.oup.com/nargab on #CoDA applications to genomics. Deadline for article submissions late autumn 2019. More details soon.

@CoDaWork2019 @IAMG_Math_Geo @socestadistica @la_UPC @ajterrassa Thank you for organizing this conference! It was great to see the awesome research happening

Closing ceremony and awards #CoDaWork2019 Best Poster: N. Stefelová Best young contribution: C. Gozzi Best Contribution: R. Tolosana-Delgado See you un two years! Thanks to all the #participants and #sponsors #sort @IAMG_Math_Geo #statmod @socestadistica @la_UPC @ajterrassa

I. Galvan presents on using principal balances to undercover GWAS / metabolite associations. #CoDaWork2019 @CoDaWork2019

And the two last #bio talks: Tom Lillhonga, #Compositional #Data Analysis of Finnish Birch Sap; and, Iván Galván, #Principal #balances for #metabolite genome-wide association analysis in the @GCATbiobank cohort #CoDaWork2019

Last #CoDaWork2019 session. A #bio session: Ionas Erb (@badlytrainedtec), #Partial #Correlations in #Compositional Data Analysis; Dennis E. te Beest, Possible pitfall of using #logratio analysis on #metagenomics data with many #zeroes;

Low abundance taxa are major drivers in PCA - maybe driven by the choice of pseudocount??

D. E. to Beest presents on issues with performing compositional methods on microbiome data of pseudocounts are used for imputation @CoDaWork2019 #CoDaWork2019

@badlytrainedtec presenting on partial correlations. It has been shown that compositional techniques out perform all other metrics for RNAseq network #CoDaWork2019 @CoDaWork2019

An example of such a biplot is given

The resulting biplots have nice interpretatable properties.

This compositional CCA falls under the class of generalized CCA originally developed by ter Braak. There are challenges with compositional data due to the singularity of the covariance matrix