Accurate single-molecule spot detection for image-based spatial transcriptomics with weakly supervised deep learning.

Image-based spatial transcriptomics methods enable transcriptome-scale gene expression measurements with spatial information but require complex, manually tuned analysis pipelines. We present Polaris, an analysis pipeline for image-based spatial transcriptomics that combines deep-learning models for cell segmentation and spot detection with a probabilistic gene decoder to quantify single-cell gene expression accurately. Polaris offers a unifying, turnkey solution for analyzing spatial transcriptomics data from multiplexed error-robust FISH (MERFISH), sequential fluorescence in situ hybridization (seqFISH), or in situ RNA sequencing (ISS) experiments. Polaris is available through the DeepCell software library (https://github.com/vanvalenlab/deepcell-spots) and https://www.deepcell.org.

Accurate single-molecule spot detection for image-based spatial transcriptomics with weakly supervised deep learning.

Image-based spatial transcriptomics methods enable transcriptome-scale gene expression measurements with spatial information but require complex, manually-tuned analysis pipelines. We present Polaris, an analysis pipeline for image-based spatial transcriptomics that combines deep learning models for cell segmentation and spot detection with a probabilistic gene decoder to quantify single-cell gene expression accurately. Polaris offers a unifying, turnkey solution for analyzing spatial transcriptomics data from MERFSIH, seqFISH, or ISS experiments. Polaris is available through the DeepCell software library (https://github.com/vanvalenlab/deepcell-spots) and https://www.deepcell.org.

A Foundation Model for Cell Segmentation.

Cells are a fundamental unit of biological organization, and identifying them in imaging data - cell segmentation - is a critical task for various cellular imaging experiments. While deep learning methods have led to substantial progress on this problem, most models in use are specialist models that work well for specific domains. Methods that have learned the general notion of "what is a cell" and can identify them across different domains of cellular imaging data have proven elusive. In this work, we present CellSAM, a foundation model for cell segmentation that generalizes across diverse cellular imaging data. CellSAM builds on top of the Segment Anything Model (SAM) by developing a prompt engineering approach for mask generation. We train an object detector, CellFinder, to automatically detect cells and prompt SAM to generate segmentations. We show that this approach allows a single model to achieve human-level performance for segmenting images of mammalian cells (in tissues and cell culture), yeast, and bacteria collected across various imaging modalities. We show that CellSAM has strong zero-shot performance and can be improved with a few examples via few-shot learning. We also show that CellSAM can unify bioimaging analysis workflows such as spatial transcriptomics and cell tracking. A deployed version of CellSAM is available at https://cellsam.deepcell.org/.

DeCOIL: Optimization of Degenerate Codon Libraries for Machine Learning-Assisted Protein Engineering.

With advances in machine learning (ML)-assisted protein engineering, models based on data, biophysics, and natural evolution are being used to propose informed libraries of protein variants to explore. Synthesizing these libraries for experimental screens is a major bottleneck, as the cost of obtaining large numbers of exact gene sequences is often prohibitive. Degenerate codon (DC) libraries are a cost-effective alternative for generating combinatorial mutagenesis libraries where mutations are targeted to a handful of amino acid sites. However, existing computational methods to optimize DC libraries to include desired protein variants are not well suited to design libraries for ML-assisted protein engineering. To address these drawbacks, we present DEgenerate Codon Optimization for Informed Libraries (DeCOIL), a generalized method that directly optimizes DC libraries to be useful for protein engineering: to sample protein variants that are likely to have both high fitness and high diversity in the sequence search space. Using computational simulations and wet-lab experiments, we demonstrate that DeCOIL is effective across two specific case studies, with the potential to be applied to many other use cases. DeCOIL offers several advantages over existing methods, as it is direct, easy to use, generalizable, and scalable. With accompanying software (https://github.com/jsunn-y/DeCOIL), DeCOIL can be readily implemented to generate desired informed libraries.

Neural-Fly enables rapid learning for agile flight in strong winds.

Executing safe and precise flight maneuvers in dynamic high-speed winds is important for the ongoing commoditization of uninhabited aerial vehicles (UAVs). However, because the relationship between various wind conditions and its effect on aircraft maneuverability is not well understood, it is challenging to design effective robot controllers using traditional control design methods. We present Neural-Fly, a learning-based approach that allows rapid online adaptation by incorporating pretrained representations through deep learning. Neural-Fly builds on two key observations that aerodynamics in different wind conditions share a common representation and that the wind-specific part lies in a low-dimensional space. To that end, Neural-Fly uses a proposed learning algorithm, domain adversarially invariant meta-learning (DAIML), to learn the shared representation, only using 12 minutes of flight data. With the learned representation as a basis, Neural-Fly then uses a composite adaptation law to update a set of linear coefficients for mixing the basis elements. When evaluated under challenging wind conditions generated with the Caltech Real Weather Wind Tunnel, with wind speeds up to 43.6 kilometers/hour (12.1 meters/second), Neural-Fly achieves precise flight control with substantially smaller tracking error than stateof-the-art nonlinear and adaptive controllers. In addition to strong empirical performance, the exponential stability of Neural-Fly results in robustness guarantees. Last, our control design extrapolates to unseen wind conditions, is shown to be effective for outdoor flights with only onboard sensors, and can transfer across drones with minimal performance degradation.

Self-Supervised Keypoint Discovery in Behavioral Videos.

We propose a method for learning the posture and structure of agents from unlabelled behavioral videos. Starting from the observation that behaving agents are generally the main sources of movement in behavioral videos, our method, Behavioral Keypoint Discovery (B-KinD), uses an encoder-decoder architecture with a geometric bottleneck to reconstruct the spatiotemporal difference between video frames. By focusing only on regions of movement, our approach works directly on input videos without requiring manual annotations. Experiments on a variety of agent types (mouse, fly, human, jellyfish, and trees) demonstrate the generality of our approach and reveal that our discovered keypoints represent semantically meaningful body parts, which achieve state-of-the-art performance on keypoint regression among self-supervised methods. Additionally, B-KinD achieve comparable performance to supervised keypoints on downstream tasks, such as behavior classification, suggesting that our method can dramatically reduce model training costs vis-a-vis supervised methods.

Informed training set design enables efficient machine learning-assisted directed protein evolution.

Directed evolution of proteins often involves a greedy optimization in which the mutation in the highest-fitness variant identified in each round of single-site mutagenesis is fixed. The efficiency of such a single-step greedy walk depends on the order in which beneficial mutations are identified-the process is path dependent. Here, we investigate and optimize a path-independent machine learning-assisted directed evolution (MLDE) protocol that allows in silico screening of full combinatorial libraries. In particular, we evaluate the importance of different protein encoding strategies, training procedures, models, and training set design strategies on MLDE outcome, finding the most important consideration to be the implementation of strategies that reduce inclusion of minimally informative "holes" (protein variants with zero or extremely low fitness) in training data. When applied to an epistatic, hole-filled, four-site combinatorial fitness landscape, our optimized protocol achieved the global fitness maximum up to 81-fold more frequently than single-step greedy optimization. A record of this paper's transparent peer review process is included in the supplemental information.

Multilabel Classification Models for the Prediction of Cross-Coupling Reaction Conditions.

Machine-learned ranking models have been developed for the prediction of substrate-specific cross-coupling reaction conditions. Data sets of published reactions were curated for Suzuki, Negishi, and C-N couplings, as well as Pauson-Khand reactions. String, descriptor, and graph encodings were tested as input representations, and models were trained to predict the set of conditions used in a reaction as a binary vector. Unique reagent dictionaries categorized by expert-crafted reaction roles were constructed for each data set, leading to context-aware predictions. We find that relational graph convolutional networks and gradient-boosting machines are very effective for this learning task, and we disclose a novel reaction-level graph attention operation in the top-performing model.

Task Programming: Learning Data Efficient Behavior Representations.

Specialized domain knowledge is often necessary to accurately annotate training sets for in-depth analysis, but can be burdensome and time-consuming to acquire from domain experts. This issue arises prominently in automated behavior analysis, in which agent movements or actions of interest are detected from video tracking data. To reduce annotation effort, we present TREBA: a method to learn annotation-sample efficient trajectory embedding for behavior analysis, based on multi-task self-supervised learning. The tasks in our method can be efficiently engineered by domain experts through a process we call "task programming", which uses programs to explicitly encode structured knowledge from domain experts. Total domain expert effort can be reduced by exchanging data annotation time for the construction of a small number of programmed tasks. We evaluate this trade-off using data from behavioral neuroscience, in which specialized domain knowledge is used to identify behaviors. We present experimental results in three datasets across two domains: mice and fruit flies. Using embeddings from TREBA, we reduce annotation burden by up to a factor of 10 without compromising accuracy compared to state-of-the-art features. Our results thus suggest that task programming and self-supervision can be an effective way to reduce annotation effort for domain experts.

Using deep reinforcement learning to reveal how the brain encodes abstract state-space representations in high-dimensional environments.

Humans possess an exceptional aptitude to efficiently make decisions from high-dimensional sensory observations. However, it is unknown how the brain compactly represents the current state of the environment to guide this process. The deep Q-network (DQN) achieves this by capturing highly nonlinear mappings from multivariate inputs to the values of potential actions. We deployed DQN as a model of brain activity and behavior in participants playing three Atari video games during fMRI. Hidden layers of DQN exhibited a striking resemblance to voxel activity in a distributed sensorimotor network, extending throughout the dorsal visual pathway into posterior parietal cortex. Neural state-space representations emerged from nonlinear transformations of the pixel space bridging perception to action and reward. These transformations reshape axes to reflect relevant high-level features and strip away information about task-irrelevant sensory features. Our findings shed light on the neural encoding of task representations for decision-making in real-world situations.

The Multi-Agent Behavior Dataset: Mouse Dyadic Social Interactions.

Multi-agent behavior modeling aims to understand the interactions that occur between agents. We present a multi-agent dataset from behavioral neuroscience, the Caltech Mouse Social Interactions (CalMS21) Dataset. Our dataset consists of trajectory data of social interactions, recorded from videos of freely behaving mice in a standard resident-intruder assay. To help accelerate behavioral studies, the CalMS21 dataset provides benchmarks to evaluate the performance of automated behavior classification methods in three settings: (1) for training on large behavioral datasets all annotated by a single annotator, (2) for style transfer to learn inter-annotator differences in behavior definitions, and (3) for learning of new behaviors of interest given limited training data. The dataset consists of 6 million frames of unlabeled tracked poses of interacting mice, as well as over 1 million frames with tracked poses and corresponding frame-level behavior annotations. The challenge of our dataset is to be able to classify behaviors accurately using both labeled and unlabeled tracking data, as well as being able to generalize to new settings.

Ferulic Acid Improves Depressive-Like Behavior in Prenatally-Stressed Offspring Rats via Anti-Inflammatory Activity and HPA Axis.

Prenatal stress (PS) can increase the risk of nervous, endocrine and metabolic diseases, and immune dysfunction. Ferulic acid (FA) is a dietary phenolic acid that has pharmacological properties, including potent anti-inflammatory action. We used male, prenatally-stressed offspring rats to investigate the anti-depressive-like effects and possible anti-inflammatory mechanism of FA. We determined the animal behaviors, and the mRNA expression and concentration of inflammatory cytokines, and HPA axis. In addition, we assessed the modulation of hippocampal nuclear factor-κB (NF-κB) activation, neuronal nitric oxide synthase (nNOS) and glucocorticoid receptors (GR) expression via western blotting and immunohistochemistry. Administration of FA (12.5, 25, and 50 mg/kg/day, i.g.) for 28 days markedly increased sucrose intake, and decreased immobility time and total number of crossings, center crossings, rearing, and grooming in the male PS offspring. FA significantly reduced IL-6, IL-1ß, and TNF-α concentration and increased IL-10 concentration in male, prenatally-stressed offspring, stimulated by the NF-κB pathway. In addition, FA inhibited interleukin-6 (IL-6), interleukin-1ß (IL-1ß), and tumor necrosis factor-α (TNF-α), and increased interleukin-10 (IL-10) mRNA and protein expression. Furthermore, FA markedly decreased the serum adrenocorticotropin (ACTH) and corticosterone concentration by the increase of GR protein expression. Taken together, this study revealed that FA has anti-depressive-like effects in male, prenatally-stressed offspring, partially due to its anti-inflammatory activity and hypothalamic-pituitary-adrenal (HPA) axis.