Thank you for visiting nature.com. You are using a browser version with limited support for CSS. To obtain
the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in
Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles
and JavaScript.
Identifying structural brain connectivity, also known as the connectome, is imperative for elucidating how neurons and neural networks process information. Existing algorithms for pruning big connectome data, however, have limited speed and memory performance for connectome-wide association studies. In this issue, Sreenivasan et al. propose a GPU-based implementation for connectome pruning called ReAl-LiFE and demonstrate its computational efficiency and utility by applying it to a wide range of datasets.
Normative modeling is considered one of the most promising avenues towards personalized medicine. The integration of multimodal, mechanistic and lifespan modeling will play an essential role, but significant challenges need to be overcome before this promise can be turned into reality.
Dr Srijan Kumar, assistant professor at Georgia Institute of Technology and a Forbes 30 Under 30 honoree in science, discusses with Nature Computational Science how he uses machine learning and data science to identify and mitigate malicious activities on online platforms, including misinformation and anti-Asian hate speech.
Characterizing the brain’s connectome at multiple scales is essential for unraveling fundamental principles of cortical information processing and how it impacts behavior. A GPU-based implementation for connectome pruning is proposed, achieving greater than 100-fold speedups over previous CPU-based implementations.
The identification of robust and generalizable biomarkers based on microbial abundance data is a challenging task. An algorithm shows an enhanced classification performance by quantifying shifts in microbial co-abundances.
Variational Monte Carlo is one of the most accurate methods to solve the many-electron Schrödinger equation, but suffers from high computational cost. A recent study uses a weight-sharing technique to accelerate the neural network-based variational Monte Carlo method, allowing accurate and effective simulations of molecules.
Biomimetic nanoparticles can form complexes with proteins. Structural descriptors have been identified to predict nanoparticle–protein complex formation and their interaction sites. These descriptors include geometrical and graph-theoretical molecular features that are universally applicable to all nanoscale macromolecules of both organic and inorganic chemistries.
Determining the origin of engineered DNA can help to foster responsible innovation within the biotechnology community. A convolutional neural network approach that learns distances between engineered DNA sequences and various labs that could have created them is used to accurately predict the lab-of-origin.
Accurate structural brain connectivity estimation is key to uncovering brain–behavior relationships. ReAl-LiFE, a GPU-accelerated approach, is applied for fast and reliable evaluation of individualized brain connectomes at scale.
An efficient algorithm is proposed to integrate large-scale microbiome datasets. This unbiased data integration method enables the identification of robust biomarkers associated with various diseases through assessing shifts of microbial network modules.
An adversarial domain translation framework is presented for scalable integration of single-cell atlases across samples, technical platforms, data modalities and species.
Weight-sharing is used to accelerate and to effectively pretrain neural network-based variational Monte Carlo methods when solving the electronic Schrödinger equation for multiple geometries.