OptoPi: An open source flexible platform for the analysis of small animal behaviour.

Behaviour is the ultimate output of neural circuit computations, and therefore its analysis is a cornerstone of neuroscience research. However, every animal and experimental paradigm requires different illumination conditions to capture and, in some cases, manipulate specific behavioural features. This means that researchers often develop, from scratch, their own solutions and experimental set-ups. Here, we present OptoPi, an open source, affordable (∼ £600), behavioural arena with accompanying multi-animal tracking software. The system features highly customisable and reproducible visible and infrared illumination and allows for optogenetic stimulation. OptoPi acquires images using a Raspberry Pi camera, features motorised LED-based illumination, Arduino control, as well as irradiance monitoring to fine-tune illumination conditions with real time feedback. Our open-source software (BioImageProcessing) can be used to simultaneously track multiple unmarked animals both in on-line and off-line modes. We demonstrate the functionality of OptoPi by recording and tracking under different illumination conditions the spontaneous behaviour of larval zebrafish as well as adult Drosophila flies and their first instar larvae, an experimental animal that due to its small size and transparency has classically been hard to track. Further, we showcase OptoPi's optogenetic capabilities through a series of experiments using transgenic Drosophila larvae.

Current Progress and Challenges in Large-Scale 3D Mitochondria Instance Segmentation.

In this paper, we present the results of the MitoEM challenge on mitochondria 3D instance segmentation from electron microscopy images, organized in conjunction with the IEEE-ISBI 2021 conference. Our benchmark dataset consists of two large-scale 3D volumes, one from human and one from rat cortex tissue, which are 1,986 times larger than previously used datasets. At the time of paper submission, 257 participants had registered for the challenge, 14 teams had submitted their results, and six teams participated in the challenge workshop. Here, we present eight top-performing approaches from the challenge participants, along with our own baseline strategies. Posterior to the challenge, annotation errors in the ground truth were corrected without altering the final ranking. Additionally, we present a retrospective evaluation of the scoring system which revealed that: 1) challenge metric was permissive with the false positive predictions; and 2) size-based grouping of instances did not correctly categorize mitochondria of interest. Thus, we propose a new scoring system that better reflects the correctness of the segmentation results. Although several of the top methods are compared favorably to our own baselines, substantial errors remain unsolved for mitochondria with challenging morphologies. Thus, the challenge remains open for submission and automatic evaluation, with all volumes available for download.

SARS-CoV-2 detection by a clinical diagnostic RT-LAMP assay.

The ongoing pandemic of SARS-CoV-2 calls for rapid and cost-effective methods to accurately identify infected individuals. The vast majority of patient samples is assessed for viral RNA presence by RT-qPCR. Our biomedical research institute, in collaboration between partner hospitals and an accredited clinical diagnostic laboratory, established a diagnostic testing pipeline that has reported on more than 252,000 RT-qPCR results since its commencement at the beginning of April 2020. However, due to ongoing demand and competition for critical resources, alternative testing strategies were sought. In this work, we present a clinically-validated procedure for high-throughput SARS-CoV-2 detection by RT-LAMP in 25 minutes that is robust, reliable, repeatable, sensitive, specific, and inexpensive.

Deep learning for automatic segmentation of the nuclear envelope in electron microscopy data, trained with volunteer segmentations.

Advancements in volume electron microscopy mean it is now possible to generate thousands of serial images at nanometre resolution overnight, yet the gold standard approach for data analysis remains manual segmentation by an expert microscopist, resulting in a critical research bottleneck. Although some machine learning approaches exist in this domain, we remain far from realizing the aspiration of a highly accurate, yet generic, automated analysis approach, with a major obstacle being lack of sufficient high-quality ground-truth data. To address this, we developed a novel citizen science project, Etch a Cell, to enable volunteers to manually segment the nuclear envelope (NE) of HeLa cells imaged with serial blockface scanning electron microscopy. We present our approach for aggregating multiple volunteer annotations to generate a high-quality consensus segmentation and demonstrate that data produced exclusively by volunteers can be used to train a highly accurate machine learning algorithm for automatic segmentation of the NE, which we share here, in addition to our archived benchmark data.