Segmentation of objects in microscopy images is required for many biomedical applications. We introduce object-centric embeddings (OCEs), which embed image patches such that the spatial offsets between patches cropped from the same object are preserved. Those learnt embeddings can be used to delineate individual objects and thus obtain instance segmentations. Here, we show theoretically that, under assumptions commonly found in microscopy images, OCEs can be learnt through a self-supervised task that predicts the spatial offset between image patches. Together, this forms an unsupervised cell instance segmentation method which we evaluate on nine diverse large-scale microscopy datasets. Segmentations obtained with our method lead to substantially improved results, compared to state-of-the-art baselines on six out of nine datasets, and perform on par on the remaining three datasets. If ground-truth annotations are available, our method serves as an excellent starting point for supervised training, reducing the required amount of ground-truth needed by one order of magnitude, thus substantially increasing the practical applicability of our method. Source code is available at github.com/funkelab/cellulus.

Representation learning in neural networks may be implemented with supervised or unsupervised algorithms, distinguished by the availability of instruction. In the sensory cortex, perceptual learning drives neural plasticity1-13, but it is not known whether this is due to supervised or unsupervised learning. Here we recorded populations of up to 90,000 neurons simultaneously from the primary visual cortex (V1) and higher visual areas (HVAs) while mice learned multiple tasks, as well as during unrewarded exposure to the same stimuli. Similar to previous studies, we found that neural changes in task mice were correlated with their behavioural learning. However, the neural changes were mostly replicated in mice with unrewarded exposure, suggesting that the changes were in fact due to unsupervised learning. The neural plasticity was highest in the medial HVAs and obeyed visual, rather than spatial, learning rules. In task mice only, we found a ramping reward-prediction signal in anterior HVAs, potentially involved in supervised learning. Our neural results predict that unsupervised learning may accelerate subsequent task learning, a prediction that we validated with behavioural experiments.

Preprint: https://www.biorxiv.org/content/early/2024/02/27/2024.02.25.581990

Background: Segmenting electron microscopy (EM) images of cellular and subcellular processes in the nervous system is a key step in many bioimaging pipelines involving classification and labeling of ultrastructures. However, fully automated techniques to segment images are often susceptible to noise and heterogeneity in EM images (e.g. different histological preparations, different organisms, different brain regions, etc.). Supervised techniques to address this problem are often helpful but require large sets of training data, which are often difficult to obtain in practice, especially across many conditions. Results: We propose a new, principled unsupervised algorithm to segment EM images using a two-step approach: edge detection via salient watersheds following by robust region merging. We performed experiments to gather EM neuroimages of two organisms (mouse and fruit fly) using different histological preparations and generated manually curated ground-truth segmentations. We compared our algorithm against several state-of- the-art unsupervised segmentation algorithms and found superior performance using two standard measures of under-and over-segmentation error. Conclusions: Our algorithm is general and may be applicable to other large-scale segmentation problems for bioimages.

Neuroendocrine systems in animals maintain organismal homeostasis and regulate stress response. Although a great deal of work has been done on the neuropeptides and hormones that are released and act on target organs in the periphery, the synaptic inputs onto these neuroendocrine outputs in the brain are less well understood. Here, we use the transmission electron microscopy reconstruction of a whole central nervous system in the larva to elucidate the sensory pathways and the interneurons that provide synaptic input to the neurosecretory cells projecting to the endocrine organs. Predicted by network modeling, we also identify a new carbon dioxide-responsive network that acts on a specific set of neurosecretory cells and that includes those expressing corazonin (Crz) and diuretic hormone 44 (Dh44) neuropeptides. Our analysis reveals a neuronal network architecture for combinatorial action based on sensory and interneuronal pathways that converge onto distinct combinations of neuroendocrine outputs.

Phylogenetic footprinting has revealed that cis-regulatory enhancers consist of conserved DNA sequence clusters (CSCs). Currently, there is no systematic approach for enhancer discovery and analysis that takes full-advantage of the sequence information within enhancer CSCs.

Reconstructing neuron morphologies from fluorescence microscope images plays a critical role in neuroscience studies. It relies on image segmentation to produce initial masks either for further processing or final results to represent neuronal morphologies. This has been a challenging step due to the variation and complexity of noisy intensity patterns in neuron images acquired from microscopes. Whereas progresses in deep learning have brought the goal of accurate segmentation much closer to reality, creating training data for producing powerful neural networks is often laborious. To overcome the difficulty of obtaining a vast number of annotated data, we propose a novel strategy of using two-stage generative models to simulate training data with voxel-level labels. Trained upon unlabeled data by optimizing a novel objective function of preserving predefined labels, the models are able to synthesize realistic 3D images with underlying voxel labels. We showed that these synthetic images could train segmentation networks to obtain even better performance than manually labeled data. To demonstrate an immediate impact of our work, we further showed that segmentation results produced by networks trained upon synthetic data could be used to improve existing neuron reconstruction methods.

The ability to specify the expression levels of exogenous genes inserted in the genomes of transgenic animals is critical for the success of a wide variety of experimental manipulations. Protein production can be regulated at the level of transcription, mRNA transport, mRNA half-life, or translation efficiency. In this report, we show that several well-characterized sequence elements derived from plant and insect viruses are able to function in Drosophila to increase the apparent translational efficiency of mRNAs by as much as 20-fold. These increases render expression levels sufficient for genetic constructs previously requiring multiple copies to be effective in single copy, including constructs expressing the temperature-sensitive inactivator of neuronal function Shibire(ts1), and for the use of cytoplasmic GFP to image the fine processes of neurons.

Capping Protein (CP) plays a central role in the creation of the Arp2/3-generated branched actin networks comprising lamellipodia and pseudopodia by virtue of its ability to cap the actin filament barbed end, which promotes Arp2/3-dependent filament nucleation and optimal branching. The highly conserved protein V-1/Myotrophin binds CP tightly in vitro to render it incapable of binding the barbed end. Here we addressed the physiological significance of this CP antagonist in Dictyostelium, which expresses a V-1 homolog that we show is very similar biochemically to mouse V-1. Consistent with previous studies of CP knockdown, overexpression of V-1 in Dictyostelium reduced the size of pseudopodia and the cortical content of Arp2/3 and induced the formation of filopodia. Importantly, these effects scaled positively with the degree of V-1 overexpression and were not seen with a V-1 mutant that cannot bind CP. V-1 is present in molar excess over CP, suggesting that it suppresses CP activity in the cytoplasm at steady state. Consistently, cells devoid of V-1, like cells overexpressing CP described previously, exhibited a significant decrease in cellular F-actin content. Moreover, V-1-null cells exhibited pronounced defects in macropinocytosis and chemotactic aggregation that were rescued by V-1, but not by the V-1 mutant. Together, these observations demonstrate that V-1 exerts significant influence in vivo on major actin-based processes via its ability to sequester CP. Finally, we present evidence that V-1's ability to sequester CP is regulated by phosphorylation, suggesting that cells may manipulate the level of active CP to tune their "actin phenotype."

The p2 progenitor domain in the ventral spinal cord gives rise to two interneuron subtypes: V2a and V2b. Delta-Notch-mediated cell-cell interactions between postmitotic immature neurons have been implicated in the segregation of neuron subtypes. However, lineage relationships between V2a and V2b neurons have not been reported. We address this issue using Tg[vsx1:GFP]zebrafish, a model system in which high GFP expression is initiated near the final stage of p2 progenitors. Cell fates were followed in progeny using time-lapse microscopy. Results indicate that the vast majority, if not all, of GFP-labeled p2 progenitors divide once to produce V2a/V2b neuron pairs,indicating that V2a and V2b neurons are generated by the asymmetric division of pair-producing progenitor cells. Together with evidence that Notch signaling is involved in the cell fate specification process, our results strongly suggest that Delta-Notch interactions between sister cells play a crucial role in the final outcome of these asymmetric divisions. This mechanism for determining cell fate is similar to asymmetric divisions that occur during Drosophila neurogenesis, where ganglion mother cells divide once to produce distinct neurons. However, unlike in Drosophila, the divisional axes of p2 progenitors in zebrafish were not fixed. We report that the terminal division of pair-producing progenitor cells in vertebrate neurogenesis can reproducibly produce two distinct neurons through a mechanism that may not depend on the orientation of the division axis.