High throughput screening determines the effects of many conditions on a given biological target. Currently, to estimate the effects of those conditions on other targets requires either strong modeling assumptions (e.g. similarities among targets) or separate screens. Ideally, data-driven experimentation could be used to learn accurate models for many conditions and targets without doing all possible experiments. We have previously described an active machine learning algorithm that can iteratively choose small sets of experiments to learn models of multiple effects. We now show that, with no prior knowledge and with liquid handling robotics and automated microscopy under its control, this learner accurately learned the effects of 48 chemical compounds on the subcellular localization of 48 proteins while performing only 29% of all possible experiments. The results represent the first practical demonstration of the utility of active learning-driven biological experimentation in which the set of possible phenotypes is unknown in advance.

The long-term goal of connecting scales in biological simulation can be facilitated by scale-agnostic methods. We demonstrate that the weighted ensemble (WE) strategy, initially developed for molecular simulations, applies effectively to spatially resolved cell-scale simulations. The WE approach runs an ensemble of parallel trajectories with assigned weights and uses a statistical resampling strategy of replicating and pruning trajectories to focus computational effort on difficult-to-sample regions. The method can also generate unbiased estimates of non-equilibrium and equilibrium observables, sometimes with significantly less aggregate computing time than would be possible using standard parallelization. Here, we use WE to orchestrate particle-based kinetic Monte Carlo simulations, which include spatial geometry (e.g., of organelles, plasma membrane) and biochemical interactions among mobile molecular species. We study a series of models exhibiting spatial, temporal and biochemical complexity and show that although WE has important limitations, it can achieve performance significantly exceeding standard parallel simulation--by orders of magnitude for some observables.

Modeling cell shape variation is critical to our understanding of cell biology. Previous work has demonstrated the utility of nonrigid image registration methods for the construction of nonparametric nuclear shape models in which pairwise deformation distances are measured between all shapes and are embedded into a low-dimensional shape space. Using these methods, we explore the relationship between cell shape and nuclear shape. We find that these are frequently dependent on each other and use this as the motivation for the development of combined cell and nuclear shape space models, extending nonparametric cell representations to multiple-component three-dimensional cellular shapes and identifying modes of joint shape variation. We learn a first-order dynamics model to predict cell and nuclear shapes, given shapes at a previous time point. We use this to determine the effects of endogenous protein tags or drugs on the shape dynamics of cell lines and show that tagged C1QBP reduces the correlation between cell and nuclear shape. To reduce the computational cost of learning these models, we demonstrate the ability to reconstruct shape spaces using a fraction of computed pairwise distances. The open-source tools provide a powerful basis for future studies of the molecular basis of cell organization.

Pattern recognition and classification of images are key challenges throughout the life sciences. We combined two approaches for large-scale classification of fluorescence microscopy images. First, using the publicly available data set from the Cell Atlas of the Human Protein Atlas (HPA), we integrated an image-classification task into a mainstream video game (EVE Online) as a mini-game, named Project Discovery. Participation by 322,006 gamers over 1 year provided nearly 33 million classifications of subcellular localization patterns, including patterns that were not previously annotated by the HPA. Second, we used deep learning to build an automated Localization Cellular Annotation Tool (Loc-CAT). This tool classifies proteins into 29 subcellular localization patterns and can deal efficiently with multi-localization proteins, performing robustly across different cell types. Combining the annotations of gamers and deep learning, we applied transfer learning to create a boosted learner that can characterize subcellular protein distribution with F1 score of 0.72. We found that engaging players of commercial computer games provided data that augmented deep learning and enabled scalable and readily improved image classification.

Resolving the spatial distribution of the human proteome at a subcellular level can greatly increase our understanding of human biology and disease. Here we present a comprehensive image-based map of subcellular protein distribution, the Cell Atlas, built by integrating transcriptomics and antibody-based immunofluorescence microscopy with validation by mass spectrometry. Mapping the in situ localization of 12,003 human proteins at a single-cell level to 30 subcellular structures enabled the definition of the proteomes of 13 major organelles. Exploration of the proteomes revealed single-cell variations in abundance or spatial distribution and localization of about half of the proteins to multiple compartments. This subcellular map can be used to refine existing protein-protein interaction networks and provides an important resource to deconvolute the highly complex architecture of the human cell.

Microscope images are information rich. In this issue of Cell, Christiansen et al. show that label-free images of cells can be used to predict fluorescent labels representing cell type, state, and organelle distribution using a deep-learning framework. This paves the way for computationally multiplexed assays derived from inexpensive label-free microscopy.

Pinpointing subcellular protein localizations from microscopy images is easy to the trained eye, but challenging to automate. Based on the Human Protein Atlas image collection, we held a competition to identify deep learning solutions to solve this task. Challenges included training on highly imbalanced classes and predicting multiple labels per image. Over 3 months, 2,172 teams participated. Despite convergence on popular networks and training techniques, there was considerable variety among the solutions. Participants applied strategies for modifying neural networks and loss functions, augmenting data and using pretrained networks. The winning models far outperformed our previous effort at multi-label classification of protein localization patterns by ~20%. These models can be used as classifiers to annotate new images, feature extractors to measure pattern similarity or pretrained networks for a wide range of biological applications.