Deep learning trained on hematoxylin and eosin tumor region of Interest predicts HER2 status and trastuzumab treatment response in HER2+ breast cancer.

The current standard of care for many patients with HER2-positive breast cancer is neoadjuvant chemotherapy in combination with anti-HER2 agents, based on HER2 amplification as detected by in situ hybridization (ISH) or protein immunohistochemistry (IHC). However, hematoxylin & eosin (H&E) tumor stains are more commonly available, and accurate prediction of HER2 status and anti-HER2 treatment response from H&E would reduce costs and increase the speed of treatment selection. Computational algorithms for H&E have been effective in predicting a variety of cancer features and clinical outcomes, including moderate success in predicting HER2 status. In this work, we present a novel convolutional neural network (CNN) approach able to predict HER2 status with increased accuracy over prior methods. We trained a CNN classifier on 188 H&E whole slide images (WSIs) manually annotated for tumor Regions of interest (ROIs) by our pathology team. Our classifier achieved an area under the curve (AUC) of 0.90 in cross-validation of slide-level HER2 status and 0.81 on an independent TCGA test set. Within slides, we observed strong agreement between pathologist annotated ROIs and blinded computational predictions of tumor regions / HER2 status. Moreover, we trained our classifier on pre-treatment samples from 187 HER2+ patients that subsequently received trastuzumab therapy. Our classifier achieved an AUC of 0.80 in a five-fold cross validation. Our work provides an H&E-based algorithm that can predict HER2 status and trastuzumab response in breast cancer at an accuracy that may benefit clinical evaluations.

Deep learning-based cross-classifications reveal conserved spatial behaviors within tumor histological images.

Histopathological images are a rich but incompletely explored data type for studying cancer. Manual inspection is time consuming, making it challenging to use for image data mining. Here we show that convolutional neural networks (CNNs) can be systematically applied across cancer types, enabling comparisons to reveal shared spatial behaviors. We develop CNN architectures to analyze 27,815 hematoxylin and eosin scanned images from The Cancer Genome Atlas for tumor/normal, cancer subtype, and mutation classification. Our CNNs are able to classify TCGA pathologist-annotated tumor/normal status of whole slide images (WSIs) in 19 cancer types with consistently high AUCs (0.995 ± 0.008), as well as subtypes with lower but significant accuracy (AUC 0.87 ± 0.1). Remarkably, tumor/normal CNNs trained on one tissue are effective in others (AUC 0.88 ± 0.11), with classifier relationships also recapitulating known adenocarcinoma, carcinoma, and developmental biology. Moreover, classifier comparisons reveal intra-slide spatial similarities, with an average tile-level correlation of 0.45 ± 0.16 between classifier pairs. Breast cancers, bladder cancers, and uterine cancers have spatial patterns that are particularly easy to detect, suggesting these cancers can be canonical types for image analysis. Patterns for TP53 mutations can also be detected, with WSI self- and cross-tissue AUCs ranging from 0.65-0.80. Finally, we comparatively evaluate CNNs on 170 breast and colon cancer images with pathologist-annotated nuclei, finding that both cellular and intercellular regions contribute to CNN accuracy. These results demonstrate the power of CNNs not only for histopathological classification, but also for cross-comparisons to reveal conserved spatial behaviors across tumors.

Machine learning guided association of adverse drug reactions with in vitro target-based pharmacology.

BACKGROUND: Adverse drug reactions (ADRs) are one of the leading causes of morbidity and mortality in health care. Understanding which drug targets are linked to ADRs can lead to the development of safer medicines. METHODS: Here, we analyse in vitro secondary pharmacology of common (off) targets for 2134 marketed drugs. To associate these drugs with human ADRs, we utilized FDA Adverse Event Reports and developed random forest models that predict ADR occurrences from in vitro pharmacological profiles. FINDINGS: By evaluating Gini importance scores of model features, we identify 221 target-ADR associations, which co-occur in PubMed abstracts to a greater extent than expected by chance. Amongst these are established relations, such as the association of in vitro hERG binding with cardiac arrhythmias, which further validate our machine learning approach. Evidence on bile acid metabolism supports our identification of associations between the Bile Salt Export Pump and renal, thyroid, lipid metabolism, respiratory tract and central nervous system disorders. Unexpectedly, our model suggests PDE3 is associated with 40 ADRs. INTERPRETATION: These associations provide a comprehensive resource to support drug development and human biology studies. FUNDING: This study was not supported by any formal funding bodies.

ModEx: A text mining system for extracting mode of regulation of transcription factor-gene regulatory interaction.

BACKGROUND: Transcription factors (TFs) are proteins that are fundamental to transcription and regulation of gene expression. Each TF may regulate multiple genes and each gene may be regulated by multiple TFs. TFs can act as either activator or repressor of gene expression. This complex network of interactions between TFs and genes underlies many developmental and biological processes and is implicated in several human diseases such as cancer. Hence deciphering the network of TF-gene interactions with information on mode of regulation (activation vs. repression) is an important step toward understanding the regulatory pathways that underlie complex traits. There are many experimental, computational, and manually curated databases of TF-gene interactions. In particular, high-throughput ChIP-Seq datasets provide a large-scale map or transcriptional regulatory interactions. However, these interactions are not annotated with information on context and mode of regulation. Such information is crucial to gain a global picture of gene regulatory mechanisms and can aid in developing machine learning models for applications such as biomarker discovery, prediction of response to therapy, and precision medicine. METHODS: In this work, we introduce a text-mining system to annotate ChIP-Seq derived interaction with such meta data through mining PubMed articles. We evaluate the performance of our system using gold standard small scale manually curated databases. RESULTS: Our results show that the method is able to accurately extract mode of regulation with F-score 0.77 on TRRUST curated interaction and F-score 0.96 on intersection of TRUSST and ChIP-network. We provide a HTTP REST API for our code to facilitate usage. Availibility: Source code and datasets are available for download on GitHub: https://github.com/samanfrm/modex.

Causal Inference Engine: a platform for directional gene set enrichment analysis and inference of active transcriptional regulators.

Inference of active regulatory mechanisms underlying specific molecular and environmental perturbations is essential for understanding cellular response. The success of inference algorithms relies on the quality and coverage of the underlying network of regulator-gene interactions. Several commercial platforms provide large and manually curated regulatory networks and functionality to perform inference on these networks. Adaptation of such platforms for open-source academic applications has been hindered by the lack of availability of accurate, high-coverage networks of regulatory interactions and integration of efficient causal inference algorithms. In this work, we present CIE, an integrated platform for causal inference of active regulatory mechanisms form differential gene expression data. Using a regularized Gaussian Graphical Model, we construct a transcriptional regulatory network by integrating publicly available ChIP-seq experiments with gene-expression data from tissue-specific RNA-seq experiments. Our GGM approach identifies high confidence transcription factor (TF)-gene interactions and annotates the interactions with information on mode of regulation (activation vs. repression). Benchmarks against manually curated databases of TF-gene interactions show that our method can accurately detect mode of regulation. We demonstrate the ability of our platform to identify active transcriptional regulators by using controlled in vitro overexpression and stem-cell differentiation studies and utilize our method to investigate transcriptional mechanisms of fibroblast phenotypic plasticity.

Systems genetics of nonsyndromic orofacial clefting provides insights into its complex aetiology.

Nonsyndromic oral clefting (NSOC) is although one of the most common congenital disorders worldwide, its underlying molecular basis remains elusive. This process has been hindered by the overwhelmingly high level of heterogeneity observed. Given that hitherto multiple loci and genes have been associated with NSOC, and that complex diseases are usually polygenic and show a considerable level of missing heritability, we used a systems genetics approach to reconstruct the NSOC network by integrating human-based physical and regulatory interactome with whole-transcriptome microarray data. We show that the network component contains 53% (23/43) of the curated NSOC-implicated gene set and displays a highly significant propinquity (P < 0.0001) between genes implicated at the genomic level and those differentially expressed at the transcriptome level. In addition, we identified bona fide candidate genes based on topological features and dysregulation (e.g. ANGPTL4), and similarly prioritised genes at GWA loci (e.g. MYC and CREBBP), thus providing further insight into the underlying heterogeneity of NSOC. Gene ontology analysis results were consistent with the NSOC network being associated with embryonic organ morphogenesis and also hinted at an aetiological overlap between NSOC and cancer. We therefore recommend this approach to be applied to other heterogeneous complex diseases to not only provide a molecular framework to unify genes which may seem as disparate entities linked to the same disease, but to also predict and prioritise candidate genes for further validation, thus addressing the missing heritability.