ICYMI, my EMBL seminar was recorded 🙏 It's not a research talk -- it's a mini course on different statistical paradigms and choosing what to estimate in HTS microbiome study. Check it out! (If you like, no pressure, srsly ❤️) www.youtube.com/watch?v=VvT3...
Most microbial ecologists understand the importance of rigorous statistical analysis for conducting a replicable study. However, formal courses in the statis...
This year, the STAMPS course on microbial ecology and microbiome analysis will be held from July 17-July 27th, 2024. This is a great course with lots of hands-on training! Deadline: March 11th. @amydwillis.bsky.social and I are the Course Directors and would be happy to answer questions!
The STAMPS course promotes dialogue and the exchange of ideas between experts in environmental and microbiome analysis and offers interdisciplinary bioinformatics and statistical training to practitio...
"What I discovered came as a genuine surprise... I saw that it would take about ten clones of me to be able to do all the things I was trying to do. I knew I was trying to do too much... but I didn’t realize how extreme the situation was." 💯😻 Thanks @hormiga.bsky.social for the specific rec's too 🙏🏻
If you can't fit things into your schedule, then you don't have control over your own priorities. How to reclaim your purpose in being a scientist if you can't find time to do what you want to do?🧪
What does it mean that a half-hour spot on your calendar a few weeks away is such a rare and valuable commodity?
radEmu is here! ❤️🐦📈🔥 The StatDivLab's best differential abundance method yet! arxiv.org/abs/2402.05231
We consider the problem of estimating fold-changes in the expected value of a multivariate outcome that is observed subject to unknown sample-specific and category-specific perturbations. We are motivated by high-throughput sequencing studies of the abundance of microbial taxa, in which microbes are systematically over- and under-detected relative to their true abundances. Our log-linear model admits a partially identifiable estimand, and we establish full identifiability by imposing interpretable parameter constraints. To reduce bias and guarantee the existence of parameter estimates in the presence of sparse observations, we apply an asymptotically negligible and constraint-invariant penalty to our estimating function. We develop a fast coordinate descent algorithm for estimation, and an augmented Lagrangian algorithm for estimation under null hypotheses. We construct a model-robust score test, and demonstrate valid inference even for small sample sizes and violated distributional assumptions. The flexibility of the approach and comparisons to related methods are illustrated via a meta-analysis of microbial associations with colorectal cancer.
Reminds me of one of my favorite poems, Academic Indian Job Description. www.mangoes-and-bullets.org/en/academic-...
Exciting 🌽corncob🌽 #microbiome#dataanalysis update! corncob now accepts any positive W's, not just counts! 🔢➡️#️⃣ We hope this will be useful for folks with (eg) coverages from shotgun data 🥳, or even proportion data🫣
I'm still moving over from the other thing 🔥💙, so (for continuity) here's the latest...
Hi folx! I'm Amy, & my research group is the Statistical Diversity Lab! 👋🏻 We develop methods for the analysis of modern #biodiversity#microbiome#eDNA. I post recommendations for rigorous data analysis, updates to software, new methods & my (usually data-informed) opinions!