Chemosensitivity prediction by transcriptional profiling

Improving anti-cancer drug response prediction using multi-task learning on graph convolutional networks

Predicting the therapeutic effect of anti-cancer drugs on tumors based on the characteristics of tumors and patients is one of the important contents of precision oncology. Existing computational methods regard the drug response prediction problem as a classification or regression task. However, few of them consider leveraging the relationship between the two tasks. In this work, we propose a Multi-task Interaction Graph Convolutional Network (MTIGCN) for anti-cancer drug response prediction. MTIGCN first utilizes an graph convolutional network-based model to produce embeddings for both cell lines and drugs. After that, the model employs multi-task learning to predict anti-cancer drug response, which involves training the model on three different tasks simultaneously: the main task of the drug sensitive or resistant classification task and the two auxiliary tasks of regression prediction and similarity network reconstruction. By sharing parameters and optimizing the losses of different tasks simultaneously, MTIGCN enhances the feature representation and reduces overfitting. The results of the experiments on two in vitro datasets demonstrated that MTIGCN outperformed seven state-of-the-art baseline methods. Moreover, the well-trained model on the in vitro dataset GDSC exhibited good performance when applied to predict drug responses in in vivo datasets PDX and TCGA. The case study confirmed the model's ability to discover unknown drug responses in cell lines.

Predicting Drug-Disease Associations Through Similarity Network Fusion and Multi-View Feature Projection Representation

Predicting drug-disease associations (DDAs) through computational methods has become a prevalent trend in drug development because of their high efficiency and low cost. Existing methods usually focus on constructing heterogeneous networks by collecting multiple data resources to improve prediction ability. However, potential association possibilities of numerous unconfirmed drug-related or disease-related pairs are not sufficiently considered. In this article, we propose a novel computational model to predict new DDAs. First, a heterogeneous network is constructed, including four types of nodes (drugs, targets, cell lines, diseases) and three types of edges (associations, association scores, similarities). Second, an updating and merging-based similarity network fusion method, termed UM-SF, is presented to fuse various similarity networks with diverse weights. Finally, an intermediate layer-mediated multi-view feature projection representation method, termed IM-FP, is proposed to calculate the predicted DDA scores. This method uses multiple association scores to construct multi-view drug features, then projects them into disease space through the intermediate layer, where an intermediate layer similarity constraint is designed to learn the projection matrices. Results of comparative experiments reveal the effectiveness of our innovations. Comparisons with other state-of-the-art models by the 10-fold cross-validation experiment indicate our model's advantage on AUROC and AUPR metrics. Moreover, our proposed model successfully predicted 107 novel high-ranked DDAs.

Improving drug response prediction based on two-space graph convolution

Patients with the same cancer types may present different genomic features and therefore have different drug sensitivities. Accordingly, correctly predicting patients' responses to the drugs can guide treatment decisions and improve the outcome of cancer patients. Existing computational methods leverage the graph convolution network model to aggregate features of different types of nodes in the heterogeneous network. They most fail to consider the similarity between homogeneous nodes. To this end, we propose an algorithm based on two-space graph convolutional neural networks, TSGCNN, to predict the response of anticancer drugs. TSGCNN first constructs the cell line feature space and the drug feature space and separately performs the graph convolution operation on the feature spaces to diffuse similarity information among homogeneous nodes. After that, we generate a heterogeneous network based on the known cell line and drug relationship and perform graph convolution operations on the heterogeneous network to collect the features of different types of nodes. Subsequently, the algorithm produces the final feature representations for cell lines and drugs by adding their self features, the feature space representations, and the heterogeneous space representations. Finally, we leverage the linear correlation coefficient decoder to reconstruct the cell line-drug correlation matrix for drug response prediction based on the final representations. We tested our model on the Cancer Drug Sensitivity Data (GDSC) and Cancer Cell Line Encyclopedia (CCLE) databases. The results indicate that TSGCNN shows excellent performance drug response prediction compared with other eight state-of-the-art methods.

Molecular Modes of Action of an Aqueous Nerium oleander Extract in Cancer Cells In Vitro and In Vivo

Cancer drug resistance remains a major obstacle in clinical oncology. As most anticancer drugs are of natural origin, we investigated the anticancer potential of a standardized cold-water leaf extract from Nerium oleander L., termed Breastin. The phytochemical characterization by nuclear magnetic resonance spectroscopy (NMR) and low- and high-resolution mass spectrometry revealed several monoglycosidic cardenolides as major constituents (adynerin, neritaloside, odoroside A, odoroside H, oleandrin, and vanderoside). Breastin inhibited the growth of 14 cell lines from hematopoietic tumors and 5 of 6 carcinomas. Remarkably, the cellular responsiveness of odoroside H and neritaloside was not correlated with all other classical drug resistance mechanisms, i.e., ATP-binding cassette transporters (ABCB1, ABCB5, ABCC1, ABCG2), oncogenes (EGFR, RAS), tumor suppressors (TP53, WT1), and others (GSTP1, HSP90, proliferation rate), in 59 tumor cell lines of the National Cancer Institute (NCI, USA), indicating that Breastin may indeed bypass drug resistance. COMPARE analyses with 153 anticancer agents in 74 tumor cell lines of the Oncotest panel revealed frequent correlations of Breastin with mitosis-inhibiting drugs. Using tubulin-GFP-transfected U2OS cells and confocal microscopy, it was found that the microtubule-disturbing effect of Breastin was comparable to that of the tubulin-depolymerizing drug paclitaxel. This result was verified by a tubulin polymerization assay in vitro and molecular docking in silico. Proteome profiling of 3171 proteins in the NCI panel revealed protein subsets whose expression significantly correlated with cellular responsiveness to odoroside H and neritaloside, indicating that protein expression profiles can be identified to predict the sensitivity or resistance of tumor cells to Breastin constituents. Breastin moderately inhibited breast cancer xenograft tumors in vivo. Remarkably, in contrast to what was observed with paclitaxel monotherapy, the combination of paclitaxel and Breastin prevented tumor relapse, indicating Breastin's potential for drug combination regimens.

Deep learning methods for drug response prediction in cancer: Predominant and emerging trends

Cancer claims millions of lives yearly worldwide. While many therapies have been made available in recent years, by in large cancer remains unsolved. Exploiting computational predictive models to study and treat cancer holds great promise in improving drug development and personalized design of treatment plans, ultimately suppressing tumors, alleviating suffering, and prolonging lives of patients. A wave of recent papers demonstrates promising results in predicting cancer response to drug treatments while utilizing deep learning methods. These papers investigate diverse data representations, neural network architectures, learning methodologies, and evaluations schemes. However, deciphering promising predominant and emerging trends is difficult due to the variety of explored methods and lack of standardized framework for comparing drug response prediction models. To obtain a comprehensive landscape of deep learning methods, we conducted an extensive search and analysis of deep learning models that predict the response to single drug treatments. A total of 61 deep learning-based models have been curated, and summary plots were generated. Based on the analysis, observable patterns and prevalence of methods have been revealed. This review allows to better understand the current state of the field and identify major challenges and promising solution paths.

Evolution of precision oncology-guided treatment paradigms

Cancer treatment is gradually evolving from the classical use of nonspecific cytotoxic drugs targeting generic mechanisms of cell growth and proliferation. Instead, new "patient-specific treatment paradigms" that are based on an individual patient's tumor-specific molecular features are emerging, and these include "druggable" genomic alterations such as oncogenic driver mutations, downstream activities of cancer-signaling pathways, and the expression of specific genes involved in tumorigenesis and cancer progression. This evolving landscape of making evidence-based treatment decisions forms the foundation of precision oncology, which aims to deliver "the right drug, to the right patient and at the right time". The long-term vision for this approach is to maximize the treatment efficacy while minimizing exposure to ineffective therapy and reducing co-morbidity-related side effects. Successful clinical translation and implementation of this vision have the potential to revolutionize treatment paradigms from predominantly reactive, to more evidence-based, proactive and predictive care. In this article, we review the past and current approaches in precision oncology, and describe their remarkable power and limitations. We also speculate on the evolution of newly emerging methodologies of the future that can be used to address some of the key challenges associated with the existing paradigms. This article is categorized under: Cancer > Genetics/Genomics/Epigenetics Cancer > Molecular and Cellular Physiology Cancer > Computational Models.

DNN-PNN: A parallel deep neural network model to improve anticancer drug sensitivity

With the rapid development of deep learning techniques and large-scale genomics database, it is of great potential to apply deep learning to the prediction task of anticancer drug sensitivity, which can effectively improve the identification efficiency and accuracy of therapeutic biomarkers. In this study, we propose a parallel deep learning framework DNN-PNN, which integrates rich and heterogeneous information from gene expression and pharmaceutical chemical structure data. With the proposal of DNN-PNN, a new and more effective drug data representation strategy is introduced, that is, the correlation between features is represented by product, which alleviates the limitations of high-dimensional discrete data in deep learning. Furthermore, the framework is optimized to reduce the time complexity of the model. We conducted extensive experiments on the CCLE datasets to compare DNN-PNN with its variant DNN-FM representing the traditional feature correlation model, the component DNN or PNN alone, and the common machine learning models. It is found that DNN-PNN not only has high prediction accuracy, but also has significant advantages in stability and convergence speed.

Predicting cancer drug response using parallel heterogeneous graph convolutional networks with neighborhood interactions

MOTIVATION

Due to cancer heterogeneity, the therapeutic effect may not be the same when a cohort of patients of the same cancer type receive the same treatment. The anticancer drug response prediction may help develop personalized therapy regimens to increase survival and reduce patients' expenses. Recently, graph neural network-based methods have aroused widespread interest and achieved impressive results on the drug response prediction task. However, most of them apply graph convolution to process cell line-drug bipartite graphs while ignoring the intrinsic differences between cell lines and drug nodes. Moreover, most of these methods aggregate node-wise neighbor features but fail to consider the element-wise interaction between cell lines and drugs.

RESULTS

This work proposes a neighborhood interaction (NI)-based heterogeneous graph convolution network method, namely NIHGCN, for anticancer drug response prediction in an end-to-end way. Firstly, it constructs a heterogeneous network consisting of drugs, cell lines and the known drug response information. Cell line gene expression and drug molecular fingerprints are linearly transformed and input as node attributes into an interaction model. The interaction module consists of a parallel graph convolution network layer and a NI layer, which aggregates node-level features from their neighbors through graph convolution operation and considers the element-level of interactions with their neighbors in the NI layer. Finally, the drug response predictions are made by calculating the linear correlation coefficients of feature representations of cell lines and drugs. We have conducted extensive experiments to assess the effectiveness of our model on Cancer Drug Sensitivity Data (GDSC) and Cancer Cell Line Encyclopedia (CCLE) datasets. It has achieved the best performance compared with the state-of-the-art algorithms, especially in predicting drug responses for new cell lines, new drugs and targeted drugs. Furthermore, our model that was well trained on the GDSC dataset can be successfully applied to predict samples of PDX and TCGA, which verified the transferability of our model from cell line in vitro to the datasets in vivo.

AVAILABILITY AND IMPLEMENTATION

The source code can be obtained from https://github.com/weiba/NIHGCN.

SUPPLEMENTARY INFORMATION

Supplementary data are available at Bioinformatics online.

Low-rank regression models for multiple binary responses and their applications to cancer cell-line encyclopedia data

In this paper, we study high-dimensional multivariate logistic regression models in which a common set of covariates is used to predict multiple binary outcomes simultaneously. Our work is primarily motivated from many biomedical studies with correlated multiple responses such as the cancer cell-line encyclopedia project. We assume that the underlying regression coefficient matrix is simultaneously low-rank and row-wise sparse. We propose an intuitively appealing selection and estimation framework based on marginal model likelihood, and we develop an efficient computational algorithm for inference. We establish a novel high-dimensional theory for this nonlinear multivariate regression. Our theory is general, allowing for potential correlations between the binary responses. We propose a new type of nuclear norm penalty using the smooth clipped absolute deviation, filling the gap in the related non-convex penalization literature. We theoretically demonstrate that the proposed approach improves estimation accuracy by considering multiple responses jointly through the proposed estimator when the underlying coefficient matrix is low-rank and row-wise sparse. In particular, we establish the non-asymptotic error bounds, and both rank and row support consistency of the proposed method. Moreover, we develop a consistent rule to simultaneously select the rank and row dimension of the coefficient matrix. Furthermore, we extend the proposed methods and theory to a joint Ising model, which accounts for the dependence relationships. In our analysis of both simulated data and the cancer cell line encyclopedia data, the proposed methods outperform the existing methods in better predicting responses.

Identification of Gedunin from a Phytochemical Depository as a Novel Multidrug Resistance-Bypassing Tubulin Inhibitor of Cancer Cells

The chemotherapy of tumors is frequently limited by the development of resistance and severe side effects. Phytochemicals may offer promising candidates to meet the urgent requirement for new anticancer drugs. We screened 69 phytochemicals, and focused on gedunin to analyze its molecular modes of action. Pearson test-base correlation analyses of the log₁₀IC₅₀ values of 55 tumor cell lines of the National Cancer Institute (NCI), USA, for gedunin with those of 91 standard anticancer agents revealed statistically significant relationships to all 10 tested microtubule inhibitors. Thus, we hypothesized that gedunin may be a novel microtubule inhibitor. Confocal microscopy, cell cycle measurements, and molecular docking in silico substantiated our assumption. Agglomerative cluster analyses and the heat map generation of proteomic data revealed a subset of 40 out of 3171 proteins, the expression of which significantly correlated with sensitivity or resistance for the NCI cell line panel to gedunin. This indicates the complexity of gedunin’s activity against cancer cells, underscoring the value of network pharmacological techniques for the investigation of the molecular modes of drug action. Finally, we correlated the transcriptome-wide mRNA expression of known drug resistance mechanism (ABC transporter, oncogenes, tumor suppressors) log₁₀IC₅₀ values for gedunin. We did not find significant correlations, indicating that gedunin’s anticancer activity might not be hampered by classical drug resistance mechanisms. In conclusion, gedunin is a novel microtubule-inhibiting drug candidate which is not involved in multidrug resistance mechanisms such as other clinically established mitotic spindle poisons.

Repurposing Itraconazole and Hydroxychloroquine to Target Lysosomal Homeostasis in Epithelial Ovarian Cancer

Drug repurposing is an attractive option for oncology drug development. Itraconazole is an antifungal ergosterol synthesis inhibitor that has pleiotropic actions including cholesterol antagonism, inhibition of Hedgehog and mTOR pathways. We tested a panel of 28 epithelial ovarian cancer (EOC) cell lines with itraconazole to define its spectrum of activity. To identify synthetic lethality in combination with itraconazole, a whole-genome drop-out genome-scale clustered regularly interspaced short palindromic repeats sensitivity screen in two cell lines (TOV1946 and OVCAR5) was performed. On this basis, we conducted a phase I dose-escalation study assessing the combination of itraconazole and hydroxychloroquine in patients with platinum refractory EOC (NCT03081702). We identified a wide spectrum of sensitivity to itraconazole across the EOC cell lines. Pathway analysis showed significant involvement of lysosomal compartments, the trans-golgi network and late endosomes/lysosomes; similar pathways are phenocopied by the autophagy inhibitor, chloroquine. We then demonstrated that the combination of itraconazole and chloroquine displayed Bliss defined synergy in EOC cancer cell lines. Furthermore, there was an association of cytotoxic synergy with the ability to induce functional lysosome dysfunction, by chloroquine. Within the clinical trial, 11 patients received at least one cycle of itraconazole and hydroxychloroquine. Treatment was safe and feasible with the recommended phase II dose of 300 and 600 mg twice daily, respectively. No objective responses were detected. Pharmacodynamic measurements on serial biopsies demonstrated limited pharmacodynamic impact. In vitro, itraconazole and chloroquine have synergistic activity and exert a potent antitumor effect by affecting lysosomal function. The drug combination had no clinical antitumor activity in dose escalation.

Significance

The combination of the antifungal drug itraconazole with antimalarial drug hydroxychloroquine leads to a cytotoxic lysosomal dysfunction, supporting the rational for further research on lysosomal targeting in ovarian cancer.

‘Breast Cancer Resistance Likelihood and Personalized Treatment Through Integrated Multiomics’

In recent times, enormous progress has been made in improving the diagnosis and therapeutic strategies for breast carcinoma, yet it remains the most prevalent cancer and second highest contributor to cancer-related deaths in women. Breast cancer (BC) affects one in eight females globally. In 2018 alone, 1.4 million cases were identified worldwide in postmenopausal women and 645,000 cases in premenopausal females, and this burden is constantly increasing. This shows that still a lot of efforts are required to discover therapeutic remedies for this disease. One of the major clinical complications associated with the treatment of breast carcinoma is the development of therapeutic resistance. Multidrug resistance (MDR) and consequent relapse on therapy are prevalent issues related to breast carcinoma; it is due to our incomplete understanding of the molecular mechanisms of breast carcinoma disease. Therefore, elucidating the molecular mechanisms involved in drug resistance is critical. For management of breast carcinoma, the treatment decision not only depends on the assessment of prognosis factors but also on the evaluation of pathological and clinical factors. Integrated data assessments of these multiple factors of breast carcinoma through multiomics can provide significant insight and hope for making therapeutic decisions. This omics approach is particularly helpful since it identifies the biomarkers of disease progression and treatment progress by collective characterization and quantification of pools of biological molecules within and among the cancerous cells. The scrupulous understanding of cancer and its treatment at the molecular level led to the concept of a personalized approach, which is one of the most significant advancements in modern oncology. Likewise, there are certain genetic and non-genetic tests available for BC which can help in personalized therapy. Genetically inherited risks can be screened for personal predisposition to BC, and genetic changes or variations (mutations) can also be identified to decide on the best treatment. Ultimately, further understanding of BC at the molecular level (multiomics) will define more precise choices in personalized medicine. In this review, we have summarized therapeutic resistance associated with BC and the techniques used for its management.

Kinome-Wide Profiling Identifies Human WNK3 as a Target of Cajanin Stilbene Acid from Cajanus cajan (L.) Millsp

Pigeon Pea (Cajanus cajan (L.) Millsp.) is a common food crop used in many parts of the world for nutritional purposes. One of its chemical constituents is cajanin stilbene acid (CSA), which exerts anticancer activity in vitro and in vivo. In an effort to identify molecular targets of CSA, we performed a kinome-wide approach based on the measurement of the enzymatic activities of 252 human kinases. The serine-threonine kinase WNK3 (also known as protein kinase lysine-deficient 3) was identified as the most promising target of CSA with the strongest enzymatic activity inhibition in vitro and the highest binding affinity in molecular docking in silico. The lowest binding affinity and the predicted binding constant pKi of CSA (−9.65 kcal/mol and 0.084 µM) were comparable or even better than those of the known WNK3 inhibitor PP-121 (−9.42 kcal/mol and 0.123 µM). The statistically significant association between WNK3 mRNA expression and cellular responsiveness to several clinically established anticancer drugs in a panel of 60 tumor cell lines and the prognostic value of WNK3 mRNA expression in sarcoma biopsies for the survival time of 230 patients can be taken as clues that CSA-based inhibition of WNK3 may improve treatment outcomes of cancer patients and that CSA may serve as a valuable supplement to the currently used combination therapy protocols in oncology.

Pan-Cancer Prediction of Cell-Line Drug Sensitivity Using Network-Based Methods

The development of reliable predictive models for individual cancer cell lines to identify an optimal cancer drug is a crucial step to accelerate personalized medicine, but vast differences in cancer cell lines and drug characteristics make it quite challenging to develop predictive models that result in high predictive power and explain the similarity of cell lines or drugs. Our study proposes a novel network-based methodology that breaks the problem into smaller, more interpretable problems to improve the predictive power of anti-cancer drug responses in cell lines. For the drug-sensitivity study, we used the GDSC database for 915 cell lines and 200 drugs. The theory of optimal mass transport was first used to separately cluster cell lines and drugs, using gene-expression profiles and extensive cheminformatic drug features, represented in a form of data networks. To predict cell-line specific drug responses, random forest regression modeling was separately performed for each cell-line drug cluster pair. Post-modeling biological analysis was further performed to identify potential biological correlates associated with drug responses. The network-based clustering method resulted in 30 distinct cell-line drug cluster pairs. Predictive modeling on each cell-line-drug cluster outperformed alternative computational methods in predicting drug responses. We found that among the four drugs top-ranked with respect to prediction performance, three targeted the PI3K/mTOR signaling pathway. Predictive modeling on clustered subsets of cell lines and drugs improved the prediction accuracy of cell-line specific drug responses. Post-modeling analysis identified plausible biological processes associated with drug responses.

WMMDCA: Prediction of Drug Responses by Weight-Based Modular Mapping in Cancer Cell Lines

Due to the high consumption of cost and time for experimental verification in clinical trials, drug response prediction by computational models have become important challenges. The existing drug response data in diverse cell lines enable prediction of potential sensitive associations. Here, we propose a weight-based modular mapping method, named as WMMDCA, to predict drug-cell line associations. The method fully considers the effects of drugs' chemical structural feature, and adds modular information into the network projection. Leave-one-out cross-validation was used to evaluate the predictive ability of WMMDCA, which showed the best performance among several state-of-the-art methods in not only the whole dataset but also the major tissue types of cell lines. Literature support of highly ranked potential associations was found manually, demonstrating the effectiveness of WMMDCA on drug response prediction.

Prospective clinical trial testing COXEN-based gene expression models of chemosensitivity in dogs with spontaneous osteosarcoma

BACKGROUND

This study is a prospective clinical trial in dogs with osteosarcoma testing a gene expression model (GEM) predicting the chemosensitivity of tumors to carboplatin (CARBO) or doxorubicin (DOX) developed using the COXEN method.

PATIENTS AND METHODS

Sixty dogs with appendicular osteosarcoma were enrolled in this trial. RNA isolation and gene expression profiling were conducted with 2 biopsies for 54/63 screened tumors, and with a single biopsy for 9 tumors. Resulting gene expression data were used for calculation of a COXEN score for CARBO and DOX based on a previous study showing the significance of this predictor on patient outcome utilizing retrospective data (BMC Bioinformatics 17:93). Dogs were assigned adjuvant CARBO, DOX or the combination based on the results of the COXEN score following surgical removal of the tumor via amputation and were monitored for disease progression by chest radiograph every 2 months.

RESULTS

The COXEN predictor of chemosensitivity to CARBO or DOX was not a significant predictor of progression-free interval or overall survival for the trial participants. The calculation of DOX COXEN score using gene expression data from two independent biopsies of the same tumor were highly correlated (P < 0.0001), whereas the calculated CARBO COXEN score was not (P = 0.3039).

CONCLUSION

The COXEN predictor of chemosensitivity to CARBO or DOX is not a significant predictor of outcome when utilized in this prospective study. This trial represents the first prospective trial of a GEM predictor of chemosensitivity and establishes pet dogs with cancer as viable surrogates for prospective trials of prognostic indicators.

Predicting Drug Response Based on Multi-omics Fusion and Graph Convolution

Different cancer patients may respond differently to cancer treatment due to the heterogeneity of cancer. It is an urgent task to develop an efficient computational method to identify drug responses in different cell lines, which guides us to design personalized therapy for an individual patient. Hence, we propose an end-to-end algorithm, namely MOFGCN, to predict drug response in cell lines based on Multi-Omics Fusion and Graph Convolution Network. MOFGCN first fuses multiple omics data to calculate the cell line similarity and then constructs a heterogeneous network by combining the cell line similarity, drug similarity, and the known cell line-drug associations. Secondly, it learns the latent features for cancer cell lines and drugs by performing graph convolution operations on the heterogeneous network. Finally, MOFGCN applies the linear correlation coefficient to reconstruct the cancer cell line-drug correlation matrix to predict drug sensitivity. To our knowledge, this is the first attempt to combine graph convolutional neural network and linear correlation coefficient for this significant task. We performed extensive evaluation experiments on the Genomics of Drug Sensitivity in Cancer (GDSC) and Cancer Cell Line Encyclopedia (CCLE) databases to validate MOFGCNs performance. The experimental results show that MOFGCN is superior to the state-of-the-art algorithms in predicting missing drug responses. It also leads to higher performance in predicting drug responses for new cell lines, new drugs, and targeted drugs.

Histone Methyltransferase G9a Promotes the Development of Renal Cancer through Epigenetic Silencing of Tumor Suppressor Gene SPINK5

Background

Renal cell carcinoma (RCC) accounts for approximately 2–3% of malignant tumors in adults, while clear cell renal cell carcinoma accounts for 70–85% of kidney cancer cases, with an increasing incidence worldwide. G9a is the second histone methyltransferase found in mammals, catalyzing lysine and histone methylation. It regulates gene transcription by catalyzing histone methylation and interacting with transcription factors to alter the tightness of histone-DNA binding. The main purpose of this study is to explore the role and mechanism of G9a in renal cell carcinoma.

Methods

Firstly, we investigated the expression of G9a in 80 clinical tissues and four cell lines. Then, we explored the effect of G9a-specific inhibitor UNC0638 on proliferation, apoptosis, migration, and invasion of two renal cancer cell lines (786-O, SN12C). In order to study the specific mechanism, G9a knocking down renal cancer cell line was constructed by lentivirus. Finally, we identified the downstream target genes of G9a using ChIP experiments and rescue experiments.

Results

The results showed that the specific G9a inhibitor UNC0638 significantly inhibited the proliferation, migration, and invasion of kidney cancer in vivo and in vitro; similar results were obtained after knocking down G9a. Meanwhile, we demonstrated that SPINK5 was one of the downstream target genes of G9a through ChIP assay and proved that G9a downregulate the expression of SPINK5 by methylation of H3K9me2. Therefore, targeting G9a might be a new approach to the treatment of kidney cancer.

Conclusion

G9a was upregulated in renal cancer and could promote the development of renal cancer in vitro and in vivo. Furthermore, we identified SPINK5 as one of the downstream target genes of G9a. Therefore, targeting G9a might be a new treatment for kidney cancer.

Assessing Consistency Across Functional Screening Datasets in Cancer Cells

Motivation

Many high-throughput screening studies have been carried out in cancer cell lines to identify therapeutic agents and targets. Existing consistency assessment studies only examined two datasets at a time, with conclusions based on a subset of carefully selected features rather than considering global consistency of all the data. However, poor concordance can still be observed for a large part of the data even when selected features are highly consistent.

Results

In this study, we assembled nine compound screening datasets and three functional genomics datasets. We derived direct measures of consistency as well as indirect measures of consistency based on association between functional data and copy number-adjusted gene expression data. These results have been integrated into a web application—the Functional Data Consistency Explorer (FDCE), to allow users to make queries and generate interactive visualizations so that functional data consistency can be assessed for individual features of interest.

Availability and implementation

The FDCE web tool and we have developed and the functional data consistency measures we have generated are available at https://lccl.shinyapps.io/FDCE/.

Supplementary information

Supplementary data are available at Bioinformatics online.

Deep learning assisted multi-omics integration for survival and drug-response prediction in breast cancer

Background

Survival and drug response are two highly emphasized clinical outcomes in cancer research that directs the prognosis of a cancer patient. Here, we have proposed a late multi omics integrative framework that robustly quantifies survival and drug response for breast cancer patients with a focus on the relative predictive ability of available omics datatypes. Neighborhood component analysis (NCA), a supervised feature selection algorithm selected relevant features from multi-omics datasets retrieved from The Cancer Genome Atlas (TCGA) and Genomics of Drug Sensitivity in Cancer (GDSC) databases. A Neural network framework, fed with NCA selected features, was used to develop survival and drug response prediction models for breast cancer patients. The drug response framework used regression and unsupervised clustering (K-means) to segregate samples into responders and non-responders based on their predicted IC50 values (Z-score).

Results

The survival prediction framework was highly effective in categorizing patients into risk subtypes with an accuracy of 94%. Compared to single-omics and early integration approaches, our drug response prediction models performed significantly better and were able to predict IC50 values (Z-score) with a mean square error (MSE) of 1.154 and an overall regression value of 0.92, showing a linear relationship between predicted and actual IC50 values.

Conclusion

The proposed omics integration strategy provides an effective way of extracting critical information from diverse omics data types enabling estimation of prognostic indicators. Such integrative models with high predictive power would have a significant impact and utility in precision oncology.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12864-021-07524-2.

Integrative pan cancer analysis reveals epigenomic variation in cancer type and cell specific chromatin domains

Epigenetic mechanisms contribute to the initiation and development of cancer, and epigenetic variation promotes dynamic gene expression patterns that facilitate tumor evolution and adaptation. While the NCI-60 panel represents a diverse set of human cancer cell lines that has been used to screen chemical compounds, a comprehensive epigenomic atlas of these cells has been lacking. Here, we report an integrative analysis of 60 human cancer epigenomes, representing a catalog of activating and repressive histone modifications. We identify genome-wide maps of canonical sharp and broad H3K4me3 domains at promoter regions of tumor suppressors, H3K27ac-marked conventional enhancers and super enhancers, and widespread inter-cancer and intra-cancer specific variability in H3K9me3 and H4K20me3-marked heterochromatin domains. Furthermore, we identify features of chromatin states, including chromatin state switching along chromosomes, correlation of histone modification density with genetic mutations, DNA methylation, enrichment of DNA binding motifs in regulatory regions, and gene activity and inactivity. These findings underscore the importance of integrating epigenomic maps with gene expression and genetic variation data to understand the molecular basis of human cancer. Our findings provide a resource for mining epigenomic maps of human cancer cells and for identifying epigenetic therapeutic targets.

Oligonucleotide Probe for Transcriptome in Vivo Analysis (TIVA) of Single Neurons with Minimal Background

Messenger RNA (mRNA) isolated from single cells can generate powerful biological insights, including the discovery of new cell types with unique functions as well as markers potentially predicting a cell's response to various therapeutic agents. We previously introduced an oligonucleotide-based technique for site-selective, photoinduced biotinylation and capture of mRNA within a living cell called transcriptome in vivo analysis (TIVA). Successful application of the TIVA technique hinges upon its oligonucleotide probe remaining completely inert (or "caged") to mRNA unless photoactivated. To improve the reliability of TIVA probe caging in diverse and challenging biological conditions, we applied a rational design process involving iterative modifications to the oligonucleotide construct. In this work, we discuss these design motivations and present an optimized probe with minimal background binding to mRNA prior to photolysis. We assess its caging performance through multiple in vitro assays including FRET analysis, native gel comigration, and pull down with model mRNA transcripts. Finally, we demonstrate that this improved probe can also isolate mRNA from single living neurons in brain tissue slices with excellent caging control.

Iterative sure independent ranking and screening for drug response prediction

Background

Prediction of drug response based on multi-omics data is a crucial task in the research of personalized cancer therapy.

Results

We proposed an iterative sure independent ranking and screening (ISIRS) scheme to select drug response-associated features and applied it to the Cancer Cell Line Encyclopedia (CCLE) dataset. For each drug in CCLE, we incorporated multi-omics data including copy number alterations, mutation and gene expression and selected up to 50 features using ISIRS. Then a linear regression model based on the selected features was exploited to predict the drug response. Cross validation test shows that our prediction accuracies are higher than existing methods for most drugs.

Conclusions

Our study indicates that the features selected by the marginal utility measure, which measures the conditional probability of drug responses given the feature, are helpful for drug response prediction.

Systematic Transcriptional Profiling of Responses to STAT1- and STAT3-Activating Cytokines in Different Cancer Types

Cytokines orchestrate responses to pathogens and in inflammatory processes, but they also play an important role in cancer by shaping the expression levels of cytokine response genes. Here, we conducted a large profiling study comparing miRNome and mRNA transcriptome data generated following different cytokine stimulations. Transcriptomic responses to STAT1- (IFNγ, IL-27) and STAT3-activating cytokines (IL6, OSM) were systematically compared in nine cancerous and non-neoplastic cell lines of different tissue origins (skin, liver and colon). The largest variation in our datasets was seen between cell lines of the three different tissues rather than stimuli. Notably, the variability in miRNome datasets was a lot more pronounced than in mRNA data. Our data also revealed that cells of skin, liver and colon tissues respond very differently to cytokines and that the cell signaling networks activated or silenced in response to STAT1- or STAT3-activating cytokines are specific to the tissue and the type of cytokine. However, globally, STAT1-activating cytokines had stronger effects than STAT3-inducing cytokines with most significant responses in liver cells, showing more genes upregulated and with higher fold change. A more detailed analysis of gene regulations upon cytokine stimulation in these cells provided insights into STAT1- versus STAT3-driven processes in hepatocarcinogenesis. Finally, independent component analysis revealed interconnected transcriptional networks distinct between cancer cells and their healthy counterparts.

Active learning effectively identifies a minimal set of maximally informative and asymptotically performant cytotoxic structure-activity patterns in NCI-60 cell lines

The NCI-60 cancer cell line screening panel has provided insights for development of subtype-specific chemical therapies and repurposing. By extracting chemical structure and cytotoxicity patterns, virtual screening potentially complements the availability of high-throughput assay platforms and improves bioactive compound discovery rates by computational prefiltering of candidate compound libraries. Many groups report high prediction performances in computational models of NCI-60 data when using cross-validation or similar techniques, yet prospective therapy development in novel cancers may have little to no such data and further may not have the resources to perform hit identification using large compound libraries. In contrast to bulk screening and analysis, the active learning methodology has demonstrated how to identify compounds for screening in small batches and update computational models iteratively, leading to predictive models with a minimum number of compounds, and importantly clarifying data volumes at which limits in predictive ability are achieved. Here, in replicate per-cell line experiments using 50% of data (∼20 000 compounds) as the external prediction target, predictive limits are reproducibly demonstrated at the stage of systematic selection of 10-30% of the incorporable half. The pattern was consistent across all 60 cell lines. Limits of predictability are found to be correlated to the doubling times of cell lines and the number of cellular response discontinuities (activity cliffs) present per cell line. Organization into chemical scaffolds delineated degrees of predictive challenge. These results provide key insights for strategies in developing new inhibitors in existing cell lines or for future automated therapy selection in personalized oncotherapy.

Natural tyrosine kinase inhibitors acting on the epidermal growth factor receptor: Their relevance for cancer therapy

Epidermal growth factor receptor (EGFR), also known as ErbB-1/HER-1, plays a key role in the regulation of the cell proliferation, migration, differentiation, and survival. Since the constitutive activation or overexpression of EGFR is nearly found in various cancers, the applications focused on EGFR are the most widely used in the clinical level, including the therapeutic drugs of targeting EGFR, monoclonal antibodies (mAbs) and tyrosine kinase inhibitors (TKIs).Over the past decades, the compounds from natural sources have been a productive source of novel drugs, especially in both discovery and development of anti-tumor drugs by targeting the EGFR pathways as the TKIs. This work presents a review of the compounds from natural sources as potential EGFR-TKIs involved in the regulation of cancer. Moreover, high-throughput drug screening of EGFR-TKIs from the natural compounds has also been summarized.

An Improved Anticancer Drug-Response Prediction Based on an Ensemble Method Integrating Matrix Completion and Ridge Regression

In this study, we proposed an ensemble learning method, simultaneously integrating a low-rank matrix completion model and a ridge regression model to predict anticancer drug response on cancer cell lines. The model was applied to two benchmark datasets, including the Cancer Cell Line Encyclopedia (CCLE) and the Genomics of Drug Sensitivity in Cancer (GDSC). As previous studies suggest, the dual-layer integrated cell line-drug network model was one of the best models by far and outperformed most state-of-the-art models. Thus, we performed a head-to-head comparison between the dual-layer integrated cell line-drug network model and our model by a 10-fold crossvalidation study. For the CCLE dataset, our model has a higher Pearson correlation coefficient between predicted and observed drug responses than that of the dual-layer integrated cell line-drug network model in 18 out of 23 drugs. For the GDSC dataset, our model is better in 26 out of 28 drugs in the phosphatidylinositol 3-kinase (PI3K) pathway and 26 out of 30 drugs in the extracellular signal-regulated kinase (ERK) signaling pathway, respectively. Based on the prediction results, we carried out two types of case studies, which further verified the effectiveness of the proposed model on the drug-response prediction. In addition, our model is more biologically interpretable than the compared method, since it explicitly outputs the genes involved in the prediction, which are enriched in functions, like transcription, Src homology 2/3 (SH2/3) domain, cell cycle, ATP binding, and zinc finger.

Inverse Projection Representation and Category Contribution Rate for Robust Tumor Recognition

Sparse representation based classification (SRC) methods have achieved remarkable results. SRC, however, still suffer from requiring enough training samples, insufficient use of test samples, and instability of representation. In this paper, a stable inverse projection representation based classification (IPRC) is presented to tackle these problems by effectively using test samples. An IPR is first proposed and its feasibility and stability are analyzed. A classification criterion named category contribution rate is constructed to match the IPR and complete classification. Moreover, a statistical measure is introduced to quantify the stability of representation-based classification methods. Based on the IPRC technique, a robust tumor recognition framework is presented by interpreting microarray gene expression data, where a two-stage hybrid gene selection method is introduced to select informative genes. Finally, the functional analysis of candidate's pathogenicity-related genes is given. Extensive experiments on six public tumor microarray gene expression datasets demonstrate the proposed technique is competitive with state-of-the-art methods.

Representation transfer for differentially private drug sensitivity prediction

Motivation

Human genomic datasets often contain sensitive information that limits use and sharing of the data. In particular, simple anonymization strategies fail to provide sufficient level of protection for genomic data, because the data are inherently identifiable. Differentially private machine learning can help by guaranteeing that the published results do not leak too much information about any individual data point. Recent research has reached promising results on differentially private drug sensitivity prediction using gene expression data. Differentially private learning with genomic data is challenging because it is more difficult to guarantee privacy in high dimensions. Dimensionality reduction can help, but if the dimension reduction mapping is learned from the data, then it needs to be differentially private too, which can carry a significant privacy cost. Furthermore, the selection of any hyperparameters (such as the target dimensionality) needs to also avoid leaking private information.

Results

We study an approach that uses a large public dataset of similar type to learn a compact representation for differentially private learning. We compare three representation learning methods: variational autoencoders, principal component analysis and random projection. We solve two machine learning tasks on gene expression of cancer cell lines: cancer type classification, and drug sensitivity prediction. The experiments demonstrate significant benefit from all representation learning methods with variational autoencoders providing the most accurate predictions most often. Our results significantly improve over previous state-of-the-art in accuracy of differentially private drug sensitivity prediction.

Availability and implementation

Code used in the experiments is available at https://github.com/DPBayes/dp-representation-transfer.

A profile of 8-oxo-dGTPase activities in the NCI-60 human cancer panel: Meta-analytic insight into the regulation and role of MTH1 (NUDT1) gene expression in carcinogenesis

We measured the specific 8-oxo-dGTPase activity profile of the NCI-60 panel of malignant cell lines, and MTH1 protein levels in a subset of 16 lines. Their 8-oxo-dGTPase activity was compared to twelve publicly accessible MTH1 mRNA expression data bases and their cross-consistency was analyzed. 8-oxo-dGTPase and MTH1 protein levels in these cell lines are generally, but not always, mainly determined by MTH1 mRNA expression levels. The aneuploidy number of MTH1 gene copies only slightly affects its mRNA expression levels. By using the data mining platforms Compare and CellMiner, our 8-oxo-dGTPase profile was compared to five global gene expression datasets to identify genes whose expression levels are directly or inversely associated with 8-oxo-dGTPase. We analyzed effects of SNP within MTH1 on MTH1 mRNA level and enzyme activity. Similar association analysis was performed for five microRNA expression datasets. We identified several proteins and microRNA which might be involved in the regulation of MTH1 expression and we discuss potential mechanisms. Comparison of chemical and natural products sensitivities of the NCI-60 panel suggests seven compounds which are directly or inversely associated with 8-oxo-dGTPase. We provide an integrated picture of MTH1 expression combined from eleven consistent MTH1 mRNA and our 8-oxo-dGTPase activity NCI-60 profiles.

Modular within and between score for drug response prediction in cancer cell lines

Drug response prediction in cancer cell lines is vital to discover new anticancer drugs. However, it's still a challenging task to accurately predict drug responses in cancer cell lines. In this study, we presented a novel computational approach, named as MSDRP (modular within and between score for drug response prediction), to predict drug responses in cell lines. The method is based on a constructed heterogeneous drug-cell line network with multiple information. Compared with other state-of-the-art methods, MSDRP acquired better predictive performance, and identified potential associations between drugs and cell lines, which have been confirmed by the published literature. The source code of MSDRP is freely available at https://github.com/shimingwang1994/MSDRP.git.

Deep gene selection method to select genes from microarray datasets for cancer classification

Background

Microarray datasets consist of complex and high-dimensional samples and genes, and generally the number of samples is much smaller than the number of genes. Due to this data imbalance, gene selection is a demanding task for microarray expression data analysis.

Results

The gene set selected by DGS has shown its superior performances in cancer classification. DGS has a high capability of reducing the number of genes in the original microarray datasets. The experimental comparisons with other representative and state-of-the-art gene selection methods also showed that DGS achieved the best performance in terms of the number of selected genes, classification accuracy, and computational cost.

Conclusions

We provide an efficient gene selection algorithm can select relevant genes which are significantly sensitive to the samples’ classes. With the few discriminative genes and less cost time by the proposed algorithm achieved much high prediction accuracy on several public microarray data, which in turn verifies the efficiency and effectiveness of the proposed gene selection method.

The Protein Neddylation Inhibitor MLN4924 Suppresses Patient-Derived Glioblastoma Cells via Inhibition of ERK and AKT Signaling

Glioblastoma is a highly aggressive and lethal brain tumor, with limited treatment options. Abnormal activation of the neddylation pathway is observed in glioblastoma, and the NEDD8-activating enzyme (NAE) inhibitor, MLN4924, was previously shown to be effective in glioblastoma cell line models. However, its effect has not been tested in patient-derived glioblastoma stem cells. We first analyzed public data to determine whether NEDD8 pathway proteins are important in glioblastoma development and patient survival. NAE1 and UBA3 levels increased in glioblastoma patients; high NEDD8 levels were associated with poor clinical outcomes. Immunohistochemistry results also supported this result. The effects of MLN4924 were evaluated in 4 glioblastoma cell lines and 15 patient-derived glioblastoma stem cells using high content analysis. Glioblastoma cell lines and patient-derived stem cells were highly susceptible to MLN4924, while normal human astrocytes were resistant. In addition, there were various responses in 15 patient-derived glioblastoma stem cells upon MLN4924 treatment. Genomic analyses indicated that MLN4924 sensitive cells exhibited enrichment of Extracellular Signal Regulated Kinase (ERK) and Protein kinase B (AKT, also known as PKB) signaling. We verified that MLN4924 inhibits ERK and AKT phosphorylation in MLN4924 sensitive cells. Our findings suggest that patient-derived glioblastoma stem cells in the context of ERK and AKT activation are sensitive and highly regulated by neddylation inhibition.

Drug dose and drug choice: Optimizing medical therapy for veterinary cancer

Although novel agents hold great promise for the treatment of animal neoplasia, there may be room for significant improvement in the use of currently available agents. These improvements include altered dosing schemes, novel combinations, and patient-specific dosing or selection of agents. Previous studies have identified surrogates for "individualized dose intensity,", for example, patient size, development of adverse effects, and pharmacokinetic parameters, as potential indicators of treatment efficacy in canine lymphoma, and strategies for patient-specific dose escalation are discussed. Strategies for treatment selection in individual patients include conventional histopathology, protein-based target assessment (eg, flow cytometry, immunohistochemistry, and mass spectrometry), and gene-based target assessment (gene expression profiling and targeted or global sequencing strategies). Currently available data in animal cancer evaluating these strategies are reviewed, as well as ongoing studies and suggestions for future directions.

Migration rather than proliferation transcriptomic signatures are strongly associated with breast cancer patient survival

The efficacy of prospective cancer treatments is routinely estimated by in vitro cell-line proliferation screens. However, it is unclear whether tumor aggressiveness and patient survival are influenced more by the proliferative or the migratory properties of cancer cells. To address this question, we experimentally measured proliferation and migration phenotypes across more than 40 breast cancer cell-lines. Based on the latter, we built and validated individual predictors of breast cancer proliferation and migration levels from the cells' transcriptomics. We then apply these predictors to estimate the proliferation and migration levels of more than 1000 TCGA breast cancer tumors. Reassuringly, both estimates increase with tumor's aggressiveness, as qualified by its stage, grade, and subtype. However, predicted tumor migration levels are significantly more strongly associated with patient survival than the proliferation levels. We confirmed these findings by conducting siRNA knock-down experiments on the highly migratory MDA-MB-231 cell lines and deriving gene knock-down based proliferation and migration signatures. We show that cytoskeletal drugs might be more beneficial in patients with high predicted migration levels. Taken together, these results testify to the importance of migration levels in determining patient survival.

Probabilistic modeling of personalized drug combinations from integrated chemical screen and molecular data in sarcoma

Background

Cancer patients with advanced disease routinely exhaust available clinical regimens and lack actionable genomic medicine results, leaving a large patient population without effective treatments options when their disease inevitably progresses. To address the unmet clinical need for evidence-based therapy assignment when standard clinical approaches have failed, we have developed a probabilistic computational modeling approach which integrates molecular sequencing data with functional assay data to develop patient-specific combination cancer treatments.

Methods

Tissue taken from a murine model of alveolar rhabdomyosarcoma was used to perform single agent drug screening and DNA/RNA sequencing experiments; results integrated via our computational modeling approach identified a synergistic personalized two-drug combination. Cells derived from the primary murine tumor were allografted into mouse models and used to validate the personalized two-drug combination.

Computational modeling of single agent drug screening and RNA sequencing of multiple heterogenous sites from a single patient’s epithelioid sarcoma identified a personalized two-drug combination effective across all tumor regions. The heterogeneity-consensus combination was validated in a xenograft model derived from the patient’s primary tumor.

Cell cultures derived from human and canine undifferentiated pleomorphic sarcoma were assayed by drug screen; computational modeling identified a resistance-abrogating two-drug combination common to both cell cultures. This combination was validated in vitro via a cell regrowth assay.

Results

Our computational modeling approach addresses three major challenges in personalized cancer therapy: synergistic drug combination predictions (validated in vitro and in vivo in a genetically engineered murine cancer model), identification of unifying therapeutic targets to overcome intra-tumor heterogeneity (validated in vivo in a human cancer xenograft), and mitigation of cancer cell resistance and rewiring mechanisms (validated in vitro in a human and canine cancer model).

Conclusions

These proof-of-concept studies support the use of an integrative functional approach to personalized combination therapy prediction for the population of high-risk cancer patients lacking viable clinical options and without actionable DNA sequencing-based therapy.

Electronic supplementary material

The online version of this article (10.1186/s12885-019-5681-6) contains supplementary material, which is available to authorized users.

An Interactive Resource to Probe Genetic Diversity and Estimated Ancestry in Cancer Cell Lines

Recent work points to a lack of diversity in genomics studies from genome-wide association studies to somatic (tumor) genome analyses. Yet, population-specific genetic variation has been shown to contribute to health disparities in cancer risk and outcomes. Immortalized cancer cell lines are widely used in cancer research, from mechanistic studies to drug screening. Larger collections of cancer cell lines better represent the genomic heterogeneity found in primary tumors. Yet, the genetic ancestral origin of cancer cell lines is rarely acknowledged and often unknown. Using genome-wide genotyping data from 1,393 cancer cell lines from the Catalogue of Somatic Mutations in Cancer (COSMIC) and Cancer Cell Line Encyclopedia (CCLE), we estimated the genetic ancestral origin for each cell line. Our data indicate that cancer cell line collections are not representative of the diverse ancestry and admixture characterizing human populations. We discuss the implications of genetic ancestry and diversity of cellular models for cancer research and present an interactive tool, Estimated Cell Line Ancestry (ECLA), where ancestry can be visualized with reference populations of the 1000 Genomes Project. Cancer researchers can use this resource to identify cell line models for their studies by taking ancestral origins into consideration.

Deep-Resp-Forest: A deep forest model to predict anti-cancer drug response

The identification of therapeutic biomarkers predictive of drug response is crucial in personalized medicine. A number of computational models to predict response of anti-cancer drugs have been developed as the establishment of several pharmacogenomics screening databases. In our study, we proposed a deep cascaded forest model, Deep-Resp-Forest, to classify the anti-cancer drug response as "sensitive" or "resistant". We made three contributions in this study. Firstly, diverse molecular data could be effectively integrated to provide more information than single type of data for the classification. Combination of two types of data were tested here. Secondly, two structures based on the multi-grained scanning to transform the raw features into high-dimensional feature vectors and integrate the diverse data were proposed in our study. Thirdly, the original deep and time-consuming architecture of cascade forest was improved by a feature optimization operation, which emphasized the most discriminative features across layers. We evaluated the proposed method on the Cancer Cell Line Encyclopedia (CCLE) and Genomics of Drug Sensitivity in Cancer (GDSC) data sets and then compared with the Support Vector Machine. The proposed Deep-Resp-Forest has demonstrated the promising use of deep learning and deep forest approach on the drug response prediction tasks. The R implementation for running our experiments is available athttps://github.com/RanSuLab/Deep-Resp-Forest.

Comprehensive anticancer drug response prediction based on a simple cell line-drug complex network model

Background

Accurate prediction of anticancer drug responses in cell lines is a crucial step to accomplish the precision medicine in oncology. Although many popular computational models have been proposed towards this non-trivial issue, there is still room for improving the prediction performance by combining multiple types of genome-wide molecular data.

Results

We first demonstrated an observation on the CCLE and GDSC datasets, i.e., genetically similar cell lines always exhibit higher response correlations to structurally related drugs. Based on this observation we built a cell line-drug complex network model, named CDCN model. It captures different contributions of all available cell line-drug responses through cell line similarities and drug similarities. We executed anticancer drug response prediction on CCLE and GDSC independently. The result is significantly superior to that of some existing studies. More importantly, our model could predict the response of new drug to new cell line with considerable performance. We also divided all possible cell lines into “sensitive” and “resistant” groups by their response values to a given drug, the prediction accuracy, sensitivity, specificity and goodness of fit are also very promising.

Conclusion

CDCN model is a comprehensive tool to predict anticancer drug responses. Compared with existing methods, it is able to provide more satisfactory prediction results with less computational consumption.

Electronic supplementary material

The online version of this article (10.1186/s12859-019-2608-9) contains supplementary material, which is available to authorized users.

A novel gene selection algorithm for cancer classification using microarray datasets

Background

Microarray datasets are an important medical diagnostic tool as they represent the states of a cell at the molecular level. Available microarray datasets for classifying cancer types generally have a fairly small sample size compared to the large number of genes involved. This fact is known as a curse of dimensionality, which is a challenging problem. Gene selection is a promising approach that addresses this problem and plays an important role in the development of efficient cancer classification due to the fact that only a small number of genes are related to the classification problem. Gene selection addresses many problems in microarray datasets such as reducing the number of irrelevant and noisy genes, and selecting the most related genes to improve the classification results.

Methods

An innovative Gene Selection Programming (GSP) method is proposed to select relevant genes for effective and efficient cancer classification. GSP is based on Gene Expression Programming (GEP) method with a new defined population initialization algorithm, a new fitness function definition, and improved mutation and recombination operators. . Support Vector Machine (SVM) with a linear kernel serves as a classifier of the GSP.

Results

Experimental results on ten microarray cancer datasets demonstrate that Gene Selection Programming (GSP) is effective and efficient in eliminating irrelevant and redundant genes/features from microarray datasets. The comprehensive evaluations and comparisons with other methods show that GSP gives a better compromise in terms of all three evaluation criteria, i.e., classification accuracy, number of selected genes, and computational cost. The gene set selected by GSP has shown its superior performances in cancer classification compared to those selected by the up-to-date representative gene selection methods.

Conclusion

Gene subset selected by GSP can achieve a higher classification accuracy with less processing time.

A Review of Microarray Datasets: Where to Find Them and Specific Characteristics

The advent of DNA microarray datasets has stimulated a new line of research both in bioinformatics and in machine learning. This type of data is used to collect information from tissue and cell samples regarding gene expression differences that could be useful for disease diagnosis or for distinguishing specific types of tumor. Microarray data classification is a difficult challenge for machine learning researchers due to its high number of features and the small sample sizes. This chapter is devoted to reviewing the microarray databases most frequently used in the literature. We also make the interested reader aware of the problematic of data characteristics in this domain, such as the imbalance of the data, their complexity, and the so-called dataset shift.

Circulating Tumor and Invasive Cell Gene Expression Profile Predicts Treatment Response and Survival in Pancreatic Adenocarcinoma

Previous studies have shown that pharmacogenomic modeling of circulating tumor and invasive cells (CTICs) can predict response of pancreatic ductal adenocarcinoma (PDAC) to combination chemotherapy, predominantly 5-fluorouracil-based. We hypothesized that a similar approach could be developed to predict treatment response to standard frontline gemcitabine with nab-paclitaxel (G/nab-P) chemotherapy. Gene expression profiles for responsiveness to G/nab-P were determined in cell lines and a test set of patient samples. A prospective clinical trial was conducted, enrolling 37 patients with advanced PDAC who received G/nab-P. Peripheral blood was collected prior to treatment, after two months of treatment, and at progression. The CTICs were isolated based on a phenotype of collagen invasion. The RNA was isolated, cDNA synthesized, and qPCR gene expression analyzed. Patients were most closely matched to one of three chemotherapy response templates. Circulating tumor and invasive cells' SMAD4 expression was measured serially. The CTICs were reliably isolated and profiled from peripheral blood prior to and during chemotherapy treatment. Individual patients could be matched to distinct response templates predicting differential responses to G/nab-P treatment. Progression free survival was significantly correlated to response prediction and ΔSMAD4 was significantly associated with disease progression. These findings support phenotypic profiling and ΔSMAD4 of CTICs as promising clinical tools for choosing effective therapy in advanced PDAC, and for anticipating disease progression.

Anti-cancer Drug Response Prediction Using Neighbor-Based Collaborative Filtering with Global Effect Removal

Patients of the same cancer may differ in their responses to a specific medical therapy. Identification of predictive molecular features for drug sensitivity holds the key in the era of precision medicine. Human cell lines have harbored most of the same genetic changes found in patients’ tumors and thus are widely used in the research of drug response. In this work, we formulated drug-response prediction as a recommender system problem and then adopted a neighbor-based collaborative filtering with global effect removal (NCFGER) method to estimate anti-cancer drug responses of cell lines by integrating cell-line similarity networks and drug similarity networks based on the fact that similar cell lines and similar drugs exhibit similar responses. Specifically, we removed the global effect in the available responses and shrunk the similarity score for each cell line pair as well as each drug pair. We then used the K most similar neighbors (hybrid of cell-line-oriented and drug-oriented) in the available responses to predict the unknown ones. Through 10-fold cross-validation, this approach was shown to reach accurate and reproducible outcomes of drug sensitivity. We also discussed the biological outcomes based on the newly predicted response values.

A Hybrid Interpolation Weighted Collaborative Filtering Method for Anti-cancer Drug Response Prediction

Individualized therapies ask for the most effective regimen for each patient, while the patients' response may differ from each other. However, it is impossible to clinically evaluate each patient's response due to the large population. Human cell lines have harbored most of the same genetic changes found in patients' tumors, thus are widely used to help understand initial responses of drugs. Based on the more credible assumption that similar cell lines and similar drugs exhibit similar responses, we formulated drug response prediction as a recommender system problem, and then adopted a hybrid interpolation weighted collaborative filtering (HIWCF) method to predict anti-cancer drug responses of cell lines by incorporating cell line similarity and drug similarity shown from gene expression profiles, drug chemical structure as well as drug response similarity. Specifically, we estimated the baseline based on the available responses and shrunk the similarity score for each cell line pair as well as each drug pair. The similarity scores were then shrunk and weighted by the correlation coefficients drawn from the know response between each pair. Before used to find the K most similar neighbors for further prediction, they went through the case amplification strategy to emphasize high similarity and neglect low similarity. In the last step for prediction, cell line-oriented and drug-oriented collaborative filtering models were carried out, and the average of predicted values from both models was used as the final predicted sensitivity. Through 10-fold cross validation, this approach was shown to reach accurate and reproducible outcome for those missing drug sensitivities. We also found that the drug response similarity between cell lines or drugs may play important role in the prediction. Finally, we discussed the biological outcomes based on the newly predicted response values in GDSC dataset.

Cancer transcriptome profiling at the juncture of clinical translation

Methodological breakthroughs over the past four decades have repeatedly revolutionized transcriptome profiling. Using RNA sequencing (RNA-seq), it has now become possible to sequence and quantify the transcriptional outputs of individual cells or thousands of samples. These transcriptomes provide a link between cellular phenotypes and their molecular underpinnings, such as mutations. In the context of cancer, this link represents an opportunity to dissect the complexity and heterogeneity of tumours and to discover new biomarkers or therapeutic strategies. Here, we review the rationale, methodology and translational impact of transcriptome profiling in cancer.

A 17-molecule set as a predictor of complete response to neoadjuvant chemotherapy with docetaxel, cisplatin, and 5-fluorouracil in esophageal cancer

Background

Recently, neoadjuvant chemotherapy with docetaxel/cisplatin/5-fluorouracil (NAC-DCF) was identified as a novel strong regimen with a high rate of pathological complete response (pCR) in advanced esophageal cancer in Japan. Predicting pCR will contribute to the therapeutic strategy and the prevention of surgical invasion. However, a predictor of pCR after NAC-DCF has not yet been developed. The aim of this study was to identify a novel predictor of pCR in locally advanced esophageal cancer treated with NAC-DCF.

Patients and methods

A total of 32 patients who received NAC-DCF followed by esophagectomy between June 2013 and March 2016 were enrolled in this study. We divided the patients into the following 2 groups: pCR group (9 cases) and non-pCR group (23 cases), and compared gene expressions between these groups using DNA microarray data and KeyMolnet. Subsequently, a validation study of candidate molecular expression was performed in 7 additional cases.

Results

Seventeen molecules, including transcription factor E2F, T-cell-specific transcription factor, Src (known as “proto-oncogene tyrosine-protein kinase of sarcoma”), interferon regulatory factor 1, thymidylate synthase, cyclin B, cyclin-dependent kinase (CDK) 4, CDK, caspase-1, vitamin D receptor, histone deacetylase, MAPK/ERK kinase, bcl-2-associated X protein, runt-related transcription factor 1, PR domain zinc finger protein 1, platelet-derived growth factor receptor, and interleukin 1, were identified as candidate molecules. The molecules were mainly associated with pathways, such as transcriptional regulation by SMAD, RB/E2F, and STAT. The validation study indicated that 12 of the 17 molecules (71%) matched the trends of molecular expression.

Conclusions

A 17-molecule set that predicts pCR after NAC-DCF for locally advanced esophageal cancer was identified.