FocA: A deep learning tool for reliable, near-real-time imaging focus analysis in automated cell assay pipelines.

The increasing use of automation in cellular assays and cell culture presents significant opportunities to enhance the scale and throughput of imaging assays, but to do so, reliable data quality and consistency are critical. Realizing the full potential of automation will thus require the design of robust analysis pipelines that span the entire workflow in question. Here we present FocA, a deep learning tool that, in near real-time, identifies in-focus and out-of-focus images generated on a fully automated cell biology research platform, the NYSCF Global Stem Cell Array®. The tool is trained on small patches of downsampled images to maximize computational efficiency without compromising accuracy, and optimized to make sure no sub-quality images are stored and used in downstream analyses. The tool automatically generates balanced and maximally diverse training sets to avoid bias. The resulting model correctly identifies 100% of out-of-focus and 98% of in-focus images in under 4 s per 96-well plate, and achieves this result even in heavily downsampled data (∼30 times smaller than native resolution). Integrating the tool into automated workflows minimizes the need for human verification as well as the collection and usage of low-quality data. FocA thus offers a solution to ensure reliable image data hygiene and improve the efficiency of automated imaging workflows using minimal computational resources.

Integrating deep learning and unbiased automated high-content screening to identify complex disease signatures in human fibroblasts.

Drug discovery for diseases such as Parkinson's disease are impeded by the lack of screenable cellular phenotypes. We present an unbiased phenotypic profiling platform that combines automated cell culture, high-content imaging, Cell Painting, and deep learning. We applied this platform to primary fibroblasts from 91 Parkinson's disease patients and matched healthy controls, creating the largest publicly available Cell Painting image dataset to date at 48 terabytes. We use fixed weights from a convolutional deep neural network trained on ImageNet to generate deep embeddings from each image and train machine learning models to detect morphological disease phenotypes. Our platform's robustness and sensitivity allow the detection of individual-specific variation with high fidelity across batches and plate layouts. Lastly, our models confidently separate LRRK2 and sporadic Parkinson's disease lines from healthy controls (receiver operating characteristic area under curve 0.79 (0.08 standard deviation)), supporting the capacity of this platform for complex disease modeling and drug screening applications.

Osteoblastic MG-63 cell differentiation, contraction, and mRNA expression in stress-relaxed 3D collagen I gels.

To investigate the molecular aspects of osteoblastic interactions with a type I collagen matrix, human osteoblast-like MG-63 cells were cultured in three-dimensional (3D) collagen I gels. MG-63 cells in collagen gels expressed higher osteocalcin mRNA levels than cells in monolayer (2D) on polystyrene surfaces. Gel contraction was assessed via releasing the collagen gels from attachment following 24 h incubation in serum free, TGF-beta1-treated, or 1,25-(OH)(2)D(3)-treated media. 10 ng/ml of TGF-beta1 was optimal for enhancing contraction and led to decreased osteocalcin mRNA levels. In contrast, 50 nM 1,25-(OH)(2)D(3) led to increased osteocalcin mRNA levels, but did not affect contraction. Furthermore, the effect of contraction on gene expression was examined by releasing a subset of gels after 24 h and assessing mRNA levels by RT-PCR. Contracting gels exhibited temporally regulated differential increases in MMP-1, MMP-3, and alpha(2) integrin mRNA levels at specific time points post release. Cytochalasin D treatment immediately following release of gels inhibited contraction in a dose-dependent manner as well as prevented upregulation of MMP-1, MMP-3, and alpha2 integrin mRNA levels in contracting gels. These results suggest that osteoblastic cells generate internal loads that may affect specific gene expression, and these changes can be altered in the presence of biomediators.