Patterns最新文献

Accurate analysis of social behaviors in animals is hindered by methodological challenges. Here, we develop a segmentation tracking and clustering system (STCS) to address two major challenges in computational neuroethology: reliable multi-animal tracking and pose estimation under complex interaction conditions and providing interpretable insights into social differences guided by genotype information. We established a comprehensive, long-term, multi-animal-tracking dataset across various experimental settings. Benchmarking STCS against state-of-the-art tracking algorithms, we demonstrated its superior efficacy in analyzing behavioral experiments and establishing a robust tracking baseline. By analyzing the behavior of mice with autism spectrum disorder (ASD) using a novel weakly supervised clustering method under both solitary and social conditions, STCS reveals potential links between social stress and motor impairments. Benefiting from its modular and web-based design, STCS allows researchers to easily integrate the latest computer vision methods, enabling comprehensive behavior analysis services over the Internet, even from a single laptop.

With the exponential rise in the prevalence of automation, trust in such technology has become more critical than ever before. Trust is confidence in a particular entity, especially in regard to the consequences they can have for the trustor, and calibrated trust is the extent to which the judgments of trust are accurate. The focus of this paper is to reevaluate the general understanding of calibrating trust in automation, update this understanding, and apply it to worker’s trust in automation in the workplace. Seminal models of trust in automation were designed for automation that was already common in workforces, where the machine’s “intelligence” (i.e., capacity for decision making, cognition, and/or understanding) was limited. Now, burgeoning automation with more human-like intelligence is intended to be more interactive with workers, serving in roles such as decision aid, assistant, or collaborative coworker. Thus, we revise “calibrated trust in automation” to include more intelligent automated systems.

Denoising diffusion probabilistic models (DDPMs) have recently been shown to accurately generate complicated data such as images, audio, or time series. Experimental and clinical neuroscience also stand to benefit from this progress, as the accurate generation of neurophysiological time series can enable or improve many neuroscientific applications. Here, we present a flexible DDPM-based method for modeling multichannel, densely sampled neurophysiological recordings. DDPMs can generate realistic synthetic data for a variety of datasets from different species and recording techniques. The generated data capture important statistics, such as frequency spectra and phase-amplitude coupling, as well as fine-grained features such as sharp wave ripples. Furthermore, data can be generated based on additional information such as experimental conditions. We demonstrate the flexibility of DDPMs in several applications, including brain-state classification and missing-data imputation. In summary, DDPMs can serve as accurate generative models of neurophysiological recordings and have broad utility in the probabilistic generation of synthetic recordings for neuroscientific applications.

Serotonin (5-HT) is crucial for regulating brain functions such as mood, sleep, and cognition. This study presents a comprehensive transcriptomic analysis of 5-HT receptors (Htrs) across ≈4 million cells in the adult mouse brain using single-cell RNA sequencing (scRNA-seq) data from the Allen Institute. We observed differential transcription patterns of all 14 Htr subtypes, revealing diverse prevalence and distribution across cell classes. Remarkably, we found that 65.84% of cells transcribe RNA of at least one Htr, with frequent co-transcription of multiple Htrs, underscoring the complexity of the 5-HT system even at the single-cell dimension. Leveraging a multiplexed error-robust fluorescence in situ hybridization (MERFISH) dataset provided by Harvard University of ≈10 million cells, we analyzed the spatial distribution of each Htr, confirming previous findings and uncovering novel transcription patterns. To aid in exploring Htr transcription, we provide an online interactive visualizer.

Mistakes in machine learning practice are commonplace and can result in loss of confidence in the findings and products of machine learning. This tutorial outlines common mistakes that occur when using machine learning and what can be done to avoid them. While it should be accessible to anyone with a basic understanding of machine learning techniques, it focuses on issues that are of particular concern within academic research, such as the need to make rigorous comparisons and reach valid conclusions. It covers five stages of the machine learning process: what to do before model building, how to reliably build models, how to robustly evaluate models, how to compare models fairly, and how to report results.

Neuroanatomy is fundamental to understanding the nervous system, particularly dendritic spines, which are vital for synaptic transmission and change in response to injury or disease. Advancements in imaging have allowed for detailed three-dimensional (3D) visualization of these structures. However, existing tools for analyzing dendritic spine morphology are limited. To address this, we developed an open-source virtual reality (VR) structural analysis software ecosystem (coined “VR-SASE”) that offers a powerful, intuitive approach for analyzing dendritic spines. Our validation process confirmed the method’s superior accuracy, outperforming recognized gold-standard neural reconstruction techniques. Importantly, the VR-SASE workflow automatically calculates key morphological metrics, such as dendritic spine length, volume, and surface area, and reliably replicates established datasets from published dendritic spine studies. By integrating the Neurodata Without Borders (NWB) data standard, VR-SASE datasets can be preserved/distributed through DANDI Archives, satisfying the NIH data sharing mandate.

Molecular design based on generative models, such as variational autoencoders (VAEs), has become increasingly popular in recent years due to its efficiency for exploring high-dimensional molecular space to identify molecules with desired properties. While the efficacy of the initial model strongly depends on the training data, the sampling efficiency of the model for suggesting novel molecules with enhanced properties can be further enhanced via latent space optimization (LSO). In this paper, we propose a multi-objective LSO method that can significantly enhance the performance of generative molecular design (GMD). The proposed method adopts an iterative weighted retraining approach, where the respective weights of the molecules in the training data are determined by their Pareto efficiency. We demonstrate that our multi-objective GMD LSO method can significantly improve the performance of GMD for jointly optimizing multiple molecular properties.

The digital twin (DT) is a concept widely used in industry to create digital replicas of physical objects or systems. The dynamic, bi-directional link between the physical entity and its digital counterpart enables a real-time update of the digital entity. It can predict perturbations related to the physical object’s function. The obvious applications of DTs in healthcare and medicine are extremely attractive prospects that have the potential to revolutionize patient diagnosis and treatment. However, challenges including technical obstacles, biological heterogeneity, and ethical considerations make it difficult to achieve the desired goal. Advances in multi-modal deep learning methods, embodied AI agents, and the metaverse may mitigate some difficulties. Here, we discuss the basic concepts underlying DTs, the requirements for implementing DTs in medicine, and their current and potential healthcare uses. We also provide our perspective on five hallmarks for a healthcare DT system to advance research in this field.

Evolutionary-based machine learning models have emerged as a fascinating approach to mapping the landscape for protein evolution. Lian et al. demonstrated that evolution-based deep generative models, specifically variational autoencoders, can organize SH3 homologs in a hierarchical latent space, effectively distinguishing the specific Sho1^SH3 domains.

What can we do to mitigate climate change and achieve carbon neutrality for buildings? In their recent publication in Patterns, the authors proposed a modularized neural network incorporating physical priors for future building energy modeling, paving the way for scalable and reliable building energy modeling, optimization, retrofit designs, and buildings-to-grid integration. In this interview, the authors talk about incorporating fundamental heat transfer and thermodynamics knowledge into data-driven models.