Profile avatar
darniee.bsky.social
PhD student in computational toxicology @johnshopkins.bsky.social
6 posts 432 followers 76 following
Regular Contributor

**New Lancet DH paper** "Importance of sample size on the quality & utility of AI-based prediction models for healthcare" - for broad audience - explains why inadequate SS harms #AI model training, evaluation & performance - pushback to claims SS irrelevant to AI research πŸ‘‡ tinyurl.com/yrje52fn

We are thrilled to announce the 2025 Leading Edge Fellows! 40 outstanding postdoctoral fellows doing pioneering research in a wide range of biological and biomedical disciplines. Learn more about these exceptional scientists: www.leadingedgesymposium.org/fellows

Warm congrats to our PhD graduates.

This #RDKit blog post builds on the earlier one about chemical words; this time I'm using Morgan fingerprint bits as the "words". greglandrum.github.io/rdkit-blog/p...

ChatGPT can now handle SMILES strings πŸ§ͺβš›οΈ

This has been a long journey, but it's now out in final form. @kjamali.bsky.social academic.oup.com/bioinformati...

The European Research Council (ERC) decision to increase its support for leading researchers moving to Europe has today been formally adopted πŸ§ͺ erc.europa.eu/news-events/...

Came close to tears while making my summer writing plan. From awe at @mirya.bsky.social writing 3 books in one year, but also because I am *losing it* over the idea of finishing my book ms. this summer. Objectively I'm close, but how can I get to DONE with a 4 wk field trip and 3 wks vaca*?

With Nobel Prize laureates occasionally straying outside their applicability domains, I thought it might be helpful to re-post Expertitis #DrugDiscovery πŸ§ͺ fbdd-lit.blogspot.com/2015/04/expe...

The layout of the mobile surgery unit in New Zealand πŸš‘ #MedSky #BlueSly #DocSky

The best sign was β€œIt’s transGENIC mice, Idiot!”

At @standupforscience.bsky.social and the signs are statistically significantly excellent

Academics from poorer socio-economic backgrounds are more likely to - not publish - have outstanding publication records - introduce more novel scientific concepts - less likely to receive recognition, as measured by citations, Nobel Prize nominations, and awards. www.nber.org/papers/w33289

We translated a validation strategy in Materials to Small molecules. A key problem in cross-validation is splitting of data. Here we propose a step-forward CV. People are used to scaffold splits, time splits etc. but these all have limitations.

For PhDs and postdocs in my network, I am doing a careers fireside chat at Johns Hopkins University on Nov 18th for the biotechnology program. This will be streamed live and can be attended remotely by anyone outside the university as well! www.eventbrite.com/e/ai-meets-b...