Deep convolutional neural networks show outstanding performance in image-based phenotype classification given that all existing phenotypes are presented during the training of the network. However, in real-world high-content screening (HCS) experiments, it is often impossible to know all phenotypes in advance. Moreover, novel phenotype discovery itself can be an HCS outcome of interest. This aspect of HCS is not yet covered by classical deep learning approaches. When presenting an image with a novel phenotype to a trained network, it fails to indicate a novelty discovery but assigns the image to a wrong phenotype. To tackle this problem and address the need for novelty detection, we use a recently developed Bayesian approach for deep neural networks called Monte Carlo (MC) dropout to define different uncertainty measures for each phenotype prediction. With real HCS data, we show that these uncertainty measures allow us to identify novel or unclear phenotypes. In addition, we also found that the MC dropout method results in a significant improvement of classification accuracy. The proposed procedure used in our HCS case study can be easily transferred to any existing network architecture and will be beneficial in terms of accuracy and novelty detection.

Deep Learning has boosted artificial intelligence over the past 5 years and is seen now as one of the major technological innovation areas, predicted to replace lots of repetitive, but complex tasks of human labor within the next decade. It is also expected to be 'game changing' for research activities in pharma and life sciences, where large sets of similar yet complex data samples are systematically analyzed. Deep learning is currently conquering formerly expert domains especially in areas requiring perception, previously not amenable to standard machine learning. A typical example is the automated analysis of images which are typically produced en-masse in many domains, e. g., in high-content screening or digital pathology. Deep learning enables to create competitive applications in so-far defined core domains of 'human intelligence'. Applications of artificial intelligence have been enabled in recent years by (i) the massive availability of data samples, collected in pharma driven drug programs (='big data') as well as (ii) deep learning algorithmic advancements and (iii) increase in compute power. Such applications are based on software frameworks with specific strengths and weaknesses. Here, we introduce typical applications and underlying frameworks for deep learning with a set of practical criteria for developing production ready solutions in life science and pharma research. Based on our own experience in successfully developing deep learning applications we provide suggestions and a baseline for selecting the most suited frameworks for a future-proof and cost-effective development.

No abstract

Drug discovery programs are moving increasingly toward phenotypic imaging assays to model disease-relevant pathways and phenotypes in vitro. These assays offer richer information than target-optimized assays by investigating multiple cellular pathways simultaneously and producing multiplexed readouts. However, extracting the desired information from complex image data poses significant challenges, preventing broad adoption of more sophisticated phenotypic assays. Deep learning-based image analysis can address these challenges by reducing the effort required to analyze large volumes of complex image data at a quality and speed adequate for routine phenotypic screening in pharmaceutical research. However, while general purpose deep learning frameworks are readily available, they are not readily applicable to images from automated microscopy. During the past 3 years, we have optimized deep learning networks for this type of data and validated the approach across diverse assays with several industry partners. From this work, we have extracted five essential design principles that we believe should guide deep learning-based analysis of high-content images and multiparameter data: (1) insightful data representation, (2) automation of training, (3) multilevel quality control, (4) knowledge embedding and transfer to new assays, and (5) enterprise integration. We report a new deep learning-based software that embodies these principles, Genedata Imagence, which allows screening scientists to reliably detect stable endpoints for primary drug response, assess toxicity and safety-relevant effects, and discover new phenotypes and compound classes. Furthermore, we show how the software retains expert knowledge from its training on a particular assay and successfully reapplies it to different, novel assays in an automated fashion.

No abstract