An AI model maps how genetic variants act in the developing brain
In autism genome data it found a burden of new regulatory variants in short-lived fetal neurons.
بقلم The desk, News · · 1 دقائق قراءة
الوسوم: genomics · neuroscience · autism · nature
هذه القصة ليست بـالعربية بعد، لذا تقرأ الإنجليزية.
- الخبر
- In autism genome data it found a burden of new regulatory variants in short-lived fetal neurons.
- لماذا يهم
- Seal, from Olga Troyanskaya's group, predicts gene expression and variant effects across 26 brain regions, 30 cell types and 7 developmental stages.
- ما تم تأكيده
- The photograph is generated for this story.
مرتبط بمصدر مذكور. غير موسوم بأنه تم التحقق منه. Nature Neuroscience: Spatiotemporal AI-based modeling of gene regulation and variant effects in the developing human brain using Seal
The news
A technical report published in Nature Neuroscience on October 15 presents Seal, an interpretable transfer-learning framework for modelling gene expression and the effects of genetic variants across 26 regions, 30 cell types and 7 developmental stages of the human brain. The authors are Yun Hao, Christopher Park, Chandra Theesfeld and Olga Troyanskaya.
Why it matters
Many risk variants for brain conditions sit outside genes and act only in certain cells at certain times. A model with that resolution helps narrow where to look.
The details
Applied to genome-wide association studies, Seal identified shared and distinct regulatory architectures across six neuropsychiatric conditions that align with their clinical trajectories, the authors write.
In whole-genome data from the Simons Simplex Collection autism cohort, Seal found a significant burden of de novo regulatory variants in transient fetal excitatory neurons, pointing to effects specific to cell type and developmental stage.
What we could not check
- Predictions are computational and need lab validation.
- We relied on the published abstract and summary.
المصادر
The photograph is generated for this story. It does not show the people named in it.