Alexander Rakitko

@alexrakitko.bsky.social

Genetic testing, population genetics, multifactor diseases, ML in HealthCare, Ph.D.

🧬Can chemotherapy response in aggressive breast cancer be predicted from the tumor “ecosystem”? New @nature.com study shows that single-cell RNAseq and spatial transcriptomics can predict response to chemotherapy in triple-negative breast cancer (model reaching AUC = 0.84).

Bitter truth or sweet AI lie? A new @Nature paper shows that making LLMs more “warm” and polite can actually make them less accurate. Warm-tuned models made +8.6% more errors on medical questions. And the effect got much worse when the user was emotionally vulnerable. www.nature.com/articles/s41...

Training language models to be warm can reduce accuracy and increase sycophancy - Nature

Experiments on five different language models show that training language models to produce warmer responses can undermine the accuracy of their output, especially when users express feelings of sadne...

nature.com

J

New metabolomic aging clock trained on 168 metabolites in UK Biobank (N=225,212)⏲️ 👉R2 up to 0.35 👉associations with frailty index, telomere length, self-rated health 👉prediction of all-cause mortality 1/3

Bild

It's the era of large-scale biobanks❗️ 🧬GWAS summary stats from meta-analyses of: 👉MVP (N=630K) 👉FinnGen (N=420K) 👉UK Biobank (N=500K) ...for 330 clinical outcomes Unparalleled resource for phenome-wide explorations of genetic variation🔗: mvp-ukbb.finngen.fi

Bild