DeepFly3D, a deep learning-based software, measures limb and appendage movements in tethered, behaving Drosophila and enables precise behavioral measurements during neural recordings, stimulation, and other biological experiments.
A new computational approach for detecting sleep waveforms reveals that the 11–15 Hz sleep 'spindle', a neural rhythm implicated in memory consolidation, co-occurs widely across cortex much more often than previously thought.
Agata Wesolowska-Andersen, Grace Zhuo Yu ... Mark I McCarthy
Combination of deep learning models trained on tissue-specific genomic data and fine-mapping approaches supports efforts to identify causal variants and mechanisms at GWAS loci.
A deep learning-based image classification pipeline unleashes automated division counting and replicative lifespan analyses of single cells growing in microfluidic cellular traps.
A deep learning-based system objectively and reliably scores the Rey–Osterrieth complex figure test, enhancing the assessment of visual memory deficits from hand-drawn images in clinical settings.
'Cellular Electron Microscopy 500,000 images' (CEM500K) is a highly heterogeneous, information-rich, non-redundant, unlabeled EM dataset curated to pre-train DL algorithms for better model generalization on EM segmentation tasks.
Charles L Bormann, Manoj Kumar Kanakasabapathy ... Hadi Shafiee
A well-trained deep learning neural network can outperform and can potentially assist expertly trained embryologists in selecting embryos based on their implantation potential, even amongst high-quality euploid blastocyst embryos.
Deep mutational scanning of homologous proteins shows conservation in allosteric mechanisms but differences in molecular details within the protein family.
Johannes Thomsen, Magnus Berg Sletfjerding ... Nikos S Hatzakis
An open-source user-friendly toolbox implementing machine learning for single-molecule FRET analysis enabling experts and non-experts to reproducibly provide dynamic structural biology insights.
