New paper on fixed and random effects variable selection in mixed survival models with Hillary Heiling, Quefeng Li, Jen Jen Yeh, Joseph Ibrahim, and Xianlu Laura Peng, to appear in Statistics in Medicine: link: lnkd.in/eZsnBQZi Arxiv link: lnkd.in/esg3QYEx R package: lnkd.in/eGqETr6m

11 years ago, I thought R was unnecessary. Now, I can’t imagine life without it. The switch wasn’t easy, but it made me a better scientist. You can read my journey here: mattkmiecik.substack.com/p/my-start-w... #rstats #DataScience #OpenScience

Sayan was an amazing scholar and so kind. That's what I remember from starting off here in biostats, was how supportive he was and eager to chat about math and biology whenever I saw him at Duke. He trained many of my peers. I will miss him greatly.

Congrats to Yusha Liu on her latest paper in @naturegenet.bsky.social! She developed a novel tool to tease out patient-specific, dataset-specific, and shared factors from multi-sample single-cell sequencing data to learn more replicable cell types in PDAC Link: pubmed.ncbi.nlm.nih.gov/39747597/

"State your facts as simply as possible, even boldly. No one wants flowers of eloquence or literary ornaments in a research article" - R. B. McKerrow

Hey @mubashariqbal.com please add: go.bsky.app/Bs5m32B

Celebrating with Naim Rashid on his promotion with tenure (and birthday) @naimurashid.bsky.social It's been amazing to have Naim as a colleague at Dana Farber and then starting our labs at UNC Biostatistics over the past 8 (!) years 🙌

Semi-supervised analysis is almost always what people really want, especially when they aren’t playacting for study section or tenure Unsupervised: we don’t know anything about it Supervised: we know everything about it Neither is common in serious research. Thanks for fighting the good fight!

Congrats to @naimurashid.bsky.social on being honored with the Grizzle award and giving an amazing lecture on improving replicability of expression signatures for prostate cancer. So well deserved! 🎉

Chief in Editor? MEGA journal?? Clinics in Oncology, which is also trademarked??? Best spam email I have received today Dear Clinics in Oncology™️ Chief in Editor, I am submitting something as we speak

“All models are wrong and yours are useless: making clinical prediction models impactful for patients” nature.com/articles/s4169… A lot of great points here! You may have heard a few of these from me over the past few years, particularly about using multiomics for clinical prediction

Work from Noor Pratap Singh and Rob @robp.bsky.social on using transcript trees for DE for power and parsimony. Really excited about these applications going forward! Also including Euphy @fennecpaaaw.bsky.social and Jason Fan

Thrilled to witness our work reach publication! Many thanks to @naimurashid.bsky.social for the invaluable guidance!

Excited to share our recent work profiling clinically-relevant mutations in CRC patients receiving SOC therapies in CALGB 80405, highlighted in @JCO_ASCO. Kudos to UNC Bios student Wancen Mu for her critical contributions to this work! Paper: tinyurl.com/yhsvntkj Editorial: tinyurl.com/2554u4yx

Happy thanksgiving from…@PeppaPigFlorida! Any yes, there is an entire theme park devoted to @peppapig, and yes, we did actually go there My idea of roast pig for dinner tonight was flatly rejected

Our work developing formal methods to handle (non)ignorable missing data in supervised #deeplearning is now online at JCGS! Great work by former @uncbiostat student David Lim, now @GSK. Was not an easy problem to tackle but David did a great job! link: shorturl.at/dflrV

Excited to share the news of my appointment to associate prof with tenure @uncbiostat! Milestones like this don’t happen alone, and are made possible by amazing mentors, colleagues, collaborators, students, and especially family Thank you for everything, and for believing in me

We have a cool new tool (in rust!) for bulk sequencing-based quantification brewing (github.com/COMBINE-lab/...). It should work well for RNA-seq quantification, metagenomic abundance estimation & related problems. Will put up quickstart guide soon for those interested in testing. Feedback welcome!

Post-#AACRPan23, a trip to @MikesPastry is a must for the fam back home Still trying to figure out how the lines work, 10 years later

Side benefit of being at #AACRPan23 in Boston: getting to celebrate stellar @uncbiostat @UNCpublichealth student Hillary Heiling's successful doctoral defense! Best of luck @dfcidatascience, they are lucky to have you!

#rstats folks: any experiences with writing grants in Rmarkdown/quarto? So far its been way smoother than writing in latex, and I am hoping that citations/bibliography management will be much more straightforward than Word (endnote...) Any tips/pitfalls to watch out for?

I'm excited to announce that the UNC-Chapel Hill Department of Biology has TWO openings for tenure track Assistant Professors using genomics-informed methods to study critical questions in Plant and/or Microbial Biology. unc.peopleadmin.com/postings/265...

A friend was at an early career researcher workshop on grant writing and heard tapping from the window. She turned to see a large vulture staring at her, later joined by three more vultures, each tapping their beaks on the window. Can anyone think of symbolism more demoralizing??

Had a great visit to Michigan biostats last week! Talked about our latest work in cancer precision medicine, replicability, and clin. implementation of algorithms. Big thanks to Veera B. and Jeremy Taylor for the invite and swag! Lots of really awesome faculty, students, and research going on at UM

Please spread the word! UNC biostatistics and UNC Genetics are recruiting a tenure-track assistant professor to engage in statistical genetics/genomics methodological and collaborative research Apply here: unc.peopleadmin.com/postings/265... Details below

In honor of arriving in Chicago for Posit::conf this week, an oldie but a goodie:

Used “where’s the beef” in three talks recently and am totally stoked. Also no one got it under 40