Application of transfer learning to predict drug-induced human in vivo gene expression changes using rat in vitro and in vivo data.

The liver is the primary site for the metabolism and detoxification of many compounds, including pharmaceuticals. Consequently, it is also the primary location for many adverse reactions. As the liver is not readily accessible for sampling in humans; rodent or cell line models are often used to evaluate potential toxic effects of a novel compound or candidate drug. However, relating the results of animal and in vitro studies to relevant clinical outcomes for the human in vivo situation still proves challenging. In this study, we incorporate principles of transfer learning within a deep artificial neural network allowing us to leverage the relative abundance of rat in vitro and in vivo exposure data from the Open TG-GATEs data set to train a model to predict the expected pattern of human in vivo gene expression following an exposure given measured human in vitro gene expression. We show that domain adaptation has been successfully achieved, with the rat and human in vitro data no longer being separable in the common latent space generated by the network. The network produces physiologically plausible predictions of human in vivo gene expression pattern following an exposure to a previously unseen compound. Moreover, we show the integration of the human in vitro data in the training of the domain adaptation network significantly improves the temporal accuracy of the predicted rat in vivo gene expression pattern following an exposure to a previously unseen compound. In this way, we demonstrate the improvements in prediction accuracy that can be achieved by combining data from distinct domains.

The role of beam density and arrangement in non-coplanar IMRT exemplified by the irradiation of brain tumors - Parallels to computed tomographic imaging.

PURPOSE: Parallels between the fields of non-coplanar IMRT and non-coplanar computed tomographic reconstruction are highlighted exemplified by the identification of qualified beam configurations for the irradiation of brain tumors. METHODS AND MATERIALS: Four types of beam configurations, i.e. a pure coplanar, a quasi-isotropic and two transitional arrangements, served to systematically examine the impact of parameters such as the sampling rate and the degree of accessibility on plan quality. The resulting set of treatment techniques was compared by means of a Pinnacle3 based retrospective planning study on 18 brain tumor cases. RESULTS AND DISCUSSION: A consistent ranking of IMRT beam constellations according to plan quality was established, which directly reflects the necessities of high-quality CT imaging. Once a sufficient dense beam sampling is secured (given by compliance to Nyquist's theorem), the quasi-isotropic (QIso) irradiation produced best treatment plans, followed by a coplanar irradiation complemented by a single orthogonal non-coplanar beam (CoPl+1). Beams evenly distributed in two orthogonal planes (2-Pl), although using larger portions of the 4π space, proved to be less favorable as the beam sequence becomes less dense. The most unfavorable technique is the pure coplanar technique (CoPl). Generally, techniques with large interbeam distance, i.e. the 2-Pl technique and, to a lesser extent, QIso, are particularly sensitive to a beam number reduction. CONCLUSIONS: Rules established for high quality non-coplanar tomographic imaging are also relevant for non-coplanar IMRT. In this regard, the degree of coverage of 4π space is less important than a sufficient dense sampling.

Use of deep learning methods to translate drug-induced gene expression changes from rat to human primary hepatocytes.

In clinical trials, animal and cell line models are often used to evaluate the potential toxic effects of a novel compound or candidate drug before progressing to human trials. However, relating the results of animal and in vitro model exposures to relevant clinical outcomes in the human in vivo system still proves challenging, relying on often putative orthologs. In recent years, multiple studies have demonstrated that the repeated dose rodent bioassay, the current gold standard in the field, lacks sufficient sensitivity and specificity in predicting toxic effects of pharmaceuticals in humans. In this study, we evaluate the potential of deep learning techniques to translate the pattern of gene expression measured following an exposure in rodents to humans, circumventing the current reliance on orthologs, and also from in vitro to in vivo experimental designs. Of the applied deep learning architectures applied in this study the convolutional neural network (CNN) and a deep artificial neural network with bottleneck architecture significantly outperform classical machine learning techniques in predicting the time series of gene expression in primary human hepatocytes given a measured time series of gene expression from primary rat hepatocytes following exposure in vitro to a previously unseen compound across multiple toxicologically relevant gene sets. With a reduction in average mean absolute error across 76 genes that have been shown to be predictive for identifying carcinogenicity from 0.0172 for a random regression forest to 0.0166 for the CNN model (p < 0.05). These deep learning architecture also perform well when applied to predict time series of in vivo gene expression given measured time series of in vitro gene expression for rats.