🔥The 9th Round of Easy Loan, Earn $40 Reward is in progress❗️ ⏰ Promotion Period: January 15th - Feburary 15th, 2025 👉 Register now and check more details at gate.io/campaigns/358

Hello 👋 everyone! Finally starting with another year of exciting 😻 events! Our two organisers Anna Quaglieri and @SoroorHediyeh will open 2019 with a workshop at Nous Group | A bigger idea of success on Producing Publication-Ready Documents 📝 in #RMarkdown. Register here meetup.com/R-Ladies-Melbo…

My final top tips for papers in #R:1. 👀 for #RmdLaTex template in the #rticles 📦 or #BiocWorkflowTools for F1000Research; 2. use the flexible-general #bookdown::pdf_document2; 3. if you are a #LaTex guru 🤓 code new journal templates 🙏! rpubs.com/annaquagli/471… …#rladies

My advice re: a “project-oriented workflow” for #rstats are maturing here: whattheyforgot.org/project-orient… Do you have generally applicable tips in the spirit of “how I work”? Write them up and let me know!

I'm going to "CAREERS IN SCIENCE: Data Science". See you there? eventbrite.com/e/careers-in-s… via Eventbrite

26 million single cells...😮😮

A 🧵 on why Seurat and Scanpy's log fold change calculations are discordant. 1/ (based on the Supplementary Notes from biorxiv.org/content/10.110…).

My first method development led by John Ouyang and Enrico Petretto. scPanel achieved 86% (12/14) accuracy in classifying dSSc patients with 18 genes. Combined with Seurat, scPanel removed batch effect from test data to enable cross-dataset prediction.

Slide by Anshul Kundaje (anshulkundaje@bluesky) showing how ROC/auROC can mislead by making a crappy classifier look like it has near-perfect performance. If researchers don't present precision, beware. Photo by Cory Gilvary. #AACR18