Search

Motivation Identifying meaningful cancer driver genes in a cohort of tumors is a challenging task in cancer genomics. Although existing studies have identified known cancer drivers, most of them focus on detecting coding drivers with mutations. It is acknowledged that non-coding drivers can regulate driver mutations to promote cancer growth. In this work, we propose a novel node importance-based network analysis (NIBNA) framework to detect coding and non-coding cancer drivers. We hypothesize that cancer drivers are crucial to the formation of community structures in cancer network, and removing them from the network greatly perturbs the network structure thereby critically affecting the functioning of the network. NIBNA detects cancer drivers using a three-step process: first, a condition-specific network is built by incorporating gene expression data and gene networks; second, the community structures in the network are estimated; and third, a centrality-based metric is applied to compute node importance. Results We apply NIBNA to the BRCA dataset, and it outperforms existing state-of-art methods in detecting coding cancer drivers. NIBNA also predicts 265 miRNA drivers, and majority of these drivers have been validated in literature. Further we apply NIBNA to detect cancer subtype-specific drivers, and several predicted drivers have been validated to be associated with cancer subtypes. Lastly, we evaluate NIBNA’s performance in detecting epithelial–mesenchymal transition drivers, and we confirmed 8 coding and 13 miRNA drivers in the list of known genes. Availability and implementation The source code can be accessed at https://github.com/mandarsc/NIBNA. Supplementary information Supplementary data are available at Bioinformatics online.

Transfer learning aims to leverage valuable information in one domain to promote the learning tasks in the other domain. Some recent studies indicated that the latent information, which has a close relationship with the high-level concepts, are more suitable for cross-domain text classification than learning raw features. To obtain more latent information existing in the latent feature space, some previous methods constructed multiple latent feature spaces. However, those methods ignored that the latent information of different latent spaces may lack the relevance for promoting the adaptability of transfer learning models,even may lead to negative knowledge transfer when there exists a glaring discrepancy among the different latent spaces. Additionally, since those methods learn the latent space distributions using a strategy of direct promotion,their computational complexity increases exponentially as the number of latent spaces increases.To tackle this challenge, this paper proposes a Multiple Groups Transfer Learning (MGTL) method. MGTL first constructs multiple different latent feature spaces and then integrates the adjacent ones that have a similar latent feature dimension into one latent space group. Along this way, multiple latent space groups can be obtained. To enhance the relevance among these latent space groups, MGTL makes the adjacent groups contain one same latent space at least. Then, different groups will have more relevance than raw latent spaces. Second, MGTL utilizes an indirect-promotion strategy to connect different latent space groups.The computational complexity of MGTL increases linearly as the number of latent space groups increases and is superior to those multiple latent space methods based on direct-promotion. In addition, an iterative algorithm is proposed to solve the optimization problem. Finally, a set of systematic experiments demonstrate that MGTL outperforms all the compared existing methods. Refereed/Peer-reviewed

Discovering causal relationships from observational data is a crucial problem and it has applications in many research areas. The PC algorithm is the state-of-the-art constraint based method for causal discovery. However, runtime of the PC algorithm, in the worst-case, is exponential to the number of nodes (variables), and thus it is inefficient when being applied to high dimensional data, e.g. gene expression datasets. On another note, the advancement of computer hardware in the last decade has resulted in the widespread availability of multi-core personal computers. There is a significant motivation for designing a parallelised PC algorithm that is suitable for personal computers and does not require end users' parallel computing knowledge beyond their competency in using the PC algorithm. In this paper, we develop parallel-PC, a fast and memory efficient PC algorithm using the parallel computing technique. We apply our method to a range of synthetic and real-world high dimensional datasets. Experimental results on a dataset from the DREAM 5 challenge show that the original PC algorithm could not produce any results after running more than 24 hours; meanwhile, our parallel-PC algorithm managed to finish within around 12 hours with a 4-core CPU computer, and less than 6 hours with a 8-core CPU computer. Furthermore, we integrate parallel-PC into a causal inference method for inferring miRNA-mRNA regulatory relationships. The experimental results show that parallel-PC helps improve both the efficiency and accuracy of the causal inference algorithm., Comment: Thuc Le, Tao Hoang, Jiuyong Li, Lin Liu, Huawen Liu, Shu Hu, "A fast PC algorithm for high dimensional causal discovery with multi-core PCs", IEEE/ACM Transactions on Computational Biology and Bioinformatics, doi:10.1109/TCBB.2016.2591526

Causal discovery is a primary focus in many fields. Various methods have been developed to mine causal relationships from observational data. Most of the methods are only capable of identifying individual causes without considering their interactions. However, in real life, many effects are due to multiple factors that interact with each other. Therefore, detecting the interactions between those causal factors is essential for understanding the real causal mechanisms. So far, there are no efficient data-driven approaches to discovering causal interactions from data, especially large data sets. In this paper, we propose a general data-driven framework and develop four algorithms instantiated from the framework to detect causal interactions, directly from data. Extensive experiments on both synthetic and real-world data have shown that the proposed framework and the algorithms can achieve high effectiveness and efficiency for causal interaction discovery. Refereed/Peer-reviewed

Motivation Unravelling cancer driver genes is important in cancer research. Although computational methods have been developed to identify cancer drivers, most of them detect cancer drivers at population level. However, two patients who have the same cancer type and receive the same treatment may have different outcomes because each patient has a different genome and their disease might be driven by different driver genes. Therefore new methods are being developed for discovering cancer drivers at individual level, but existing personalized methods only focus on coding drivers while microRNAs (miRNAs) have been shown to drive cancer progression as well. Thus, novel methods are required to discover both coding and miRNA cancer drivers at individual level. Results We propose the novel method, pDriver, to discover personalized cancer drivers. pDriver includes two stages: (i) constructing gene networks for each cancer patient and (ii) discovering cancer drivers for each patient based on the constructed gene networks. To demonstrate the effectiveness of pDriver, we have applied it to five TCGA cancer datasets and compared it with the state-of-the-art methods. The result indicates that pDriver is more effective than other methods. Furthermore, pDriver can also detect miRNA cancer drivers and most of them have been confirmed to be associated with cancer by literature. We further analyze the predicted personalized drivers for breast cancer patients and the result shows that they are significantly enriched in many GO processes and KEGG pathways involved in breast cancer. Availability and implementation pDriver is available at https://github.com/pvvhoang/pDriver. Supplementary information Supplementary data are available at Bioinformatics online.

In many applications, there is a need to predict the effect of an intervention on different individuals from data. For example, which customers are persuadable by a product promotion? which patients should be treated with a certain type of treatment? These are typical causal questions involving the effect or the change in outcomes made by an intervention. The questions cannot be answered with traditional classification methods as they only use associations to predict outcomes. For personalised marketing, these questions are often answered with uplift modelling. The objective of uplift modelling is to estimate causal effect, but its literature does not discuss when the uplift represents causal effect. Causal heterogeneity modelling can solve the problem, but its assumption of unconfoundedness is untestable in data. So practitioners need guidelines in their applications when using the methods. In this paper, we use causal classification for a set of personalised decision making problems, and differentiate it from classification. We discuss the conditions when causal classification can be resolved by uplift (and causal heterogeneity) modelling methods. We also propose a general framework for causal classification, by using off-the-shelf supervised methods for flexible implementations. Experiments have shown two instantiations of the framework work for causal classification and for uplift (causal heterogeneity) modelling, and are competitive with the other uplift (causal heterogeneity) modelling methods., International Journal of Data Science and Analytics (2021). arXiv admin note: text overlap with arXiv:1604.07212 by other authors

In the disability employment service, an emerging challenge is to recommend the skills whose improvement will increase the employment perspective most. The process of a skill being improved is called an intervention and different skills are called factors. The problem involves recommendation for outcome improvement, which requires estimating the improvement in the employment perspective, i.e., the outcome, driven by interventions on recommended factors. Currently, most recommendation systems deployed for the employment service rely on traditional recommendation models where the desired outcome instead of the degree of outcome improvement is the main goal for optimization. In this paper, we present a causality-based approach for recommending factors for intervention to achieve the largest improvement in the employment potential of disabled job seekers. It involves inferring the causal effect of interventions on the employment outcome to make recommendations for individuals. The causal interpretation of our model can justify given recommendations. We conduct a case study with our industry partner in the disability employment service. Results show that the recommended interventions could improve the employability of disabled job seekers. Experiments are also carried out with datasets in other domains to demonstrate the promise of our approach in different applications. Refereed/Peer-reviewed

Identifying the genes responsible for driving cancer is of critical importance for directing treatment. Accordingly, multiple computational tools have been developed to facilitate this task. Due to the different methods employed by these tools, different data considered by the tools, and the rapidly evolving nature of the field, the selection of an appropriate tool for cancer driver discovery is not straightforward. This survey seeks to provide a comprehensive review of the different computational methods for discovering cancer drivers. We categorise the methods into three groups; methods for single driver identification, methods for driver module identification, and methods for identifying personalised cancer drivers. In addition to providing a "one-stop" reference of these methods, by evaluating and comparing their performance, we also provide readers the information about the different capabilities of the methods in identifying biologically significant cancer drivers. The biologically relevant information identified by these tools can be seen through the enrichment of discovered cancer drivers in GO biological processes and KEGG pathways and through our identification of a small cancer-driver cohort that is capable of stratifying patient survival. Refereed/Peer-reviewed

Motivation microRNAs (miRNAs) are important gene regulators and they are involved in many biological processes, including cancer progression. Therefore, correctly identifying miRNA–mRNA interactions is a crucial task. To this end, a huge number of computational methods has been developed, but they mainly use the data at one snapshot and ignore the dynamics of a biological process. The recent development of single cell data and the booming of the exploration of cell trajectories using ‘pseudotime’ concept have inspired us to develop a pseudotime-based method to infer the miRNA–mRNA relationships characterizing a biological process by taking into account the temporal aspect of the process. Results We have developed a novel approach, called pseudotime causality, to find the causal relationships between miRNAs and mRNAs during a biological process. We have applied the proposed method to both single cell and bulk sequencing datasets for Epithelia to Mesenchymal Transition, a key process in cancer metastasis. The evaluation results show that our method significantly outperforms existing methods in finding miRNA–mRNA interactions in both single cell and bulk data. The results suggest that utilizing the pseudotemporal information from the data helps reveal the gene regulation in a biological process much better than using the static information. Availability and implementation R scripts and datasets can be found at https://github.com/AndresMCB/PTC. Supplementary information Supplementary data are available at Bioinformatics online.

MotivationPredicting cell locations is important since with the understanding of cell locations, we may estimate the function of cells and their integration with the spatial environment. Thus, the DREAM Challenge on Single Cell Transcriptomics required participants to predict the locations of single cells in the Drosophila embryo using single cell transcriptomic data.ResultsWe have developed over 50 pipelines by combining different ways of pre-processing the RNA-seq data, selecting the genes, predicting the cell locations, and validating predicted cell locations, resulting in the winning methods for two out of three sub-challenges in the competition. In this paper, we present anRpackage,SCTCwhatateam, which includes all the methods we developed and theShinyweb-application to facilitate the research on single cell spatial reconstruction. All the data and the example use cases are available in the Supplementary material.AvailabilityThe scripts of the package are available athttps://github.com/thanhbuu04/SCTCwhatateamand theShinyapplication is available athttps://github.com/pvvhoang/SCTCwhatateam-ShinyAppContactThuc.Le@unisa.edu.auSupplementary informationSupplementary data are available atBriefings in Bioinformaticsonline.

Motivation Predicting cell locations is important since with the understanding of cell locations, we may estimate the function of cells and their integration with the spatial environment. Thus, the DREAM challenge on single-cell transcriptomics required participants to predict the locations of single cells in the Drosophila embryo using single-cell transcriptomic data. Results We have developed over 50 pipelines by combining different ways of preprocessing the RNA-seq data, selecting the genes, predicting the cell locations and validating predicted cell locations, resulting in the winning methods which were ranked second in sub-challenge 1, first in sub-challenge 2 and third in sub-challenge 3. In this paper, we present an R package, SCTCwhatateam, which includes all the methods we developed and the Shiny web application to facilitate the research on single-cell spatial reconstruction. All the data and the example use cases are available in the Supplementary data.

Breast cancer prognosis is challenging due to the heterogeneity of the disease. Various computational methods using bulk RNA-seq data have been proposed for breast cancer prognosis. However, these methods suffer from limited performances or ambiguous biological relevance, as a result of the neglect of intra-tumor heterogeneity. Recently, single cell RNA-sequencing (scRNA-seq) has emerged for studying tumor heterogeneity at cellular levels. In this paper, we propose a novel method, scPrognosis, to improve breast cancer prognosis with scRNA-seq data. scPrognosis uses the scRNA-seq data of the biological process Epithelial-to-Mesenchymal Transition (EMT). It firstly infers the EMT pseudotime and a dynamic gene co-expression network, then uses an integrative model to select genes important in EMT based on their expression variation and differentiation in different stages of EMT, and their roles in the dynamic gene co-expression network. To validate and apply the selected signatures to breast cancer prognosis, we use them as the features to build a prediction model with bulk RNA-seq data. The experimental results show that scPrognosis outperforms other benchmark breast cancer prognosis methods that use bulk RNA-seq data. Moreover, the dynamic changes in the expression of the selected signature genes in EMT may provide clues to the link between EMT and clinical outcomes of breast cancer. scPrognosis will also be useful when applied to scRNA-seq datasets of different biological processes other than EMT., Author summary Various computational methods have been developed for breast cancer prognosis. However, those methods mainly use the gene expression data generated by the bulk RNA sequencing techniques, which average the expression level of a gene across different cell types. As breast cancer is a heterogenous disease, the bulk gene expression may not be the ideal resource for cancer prognosis. In this study, we propose a novel method to improve breast cancer prognosis using scRNA-seq data. The proposed method has been applied to the EMT scRNA-seq dataset for identifying breast cancer signatures for prognosis. In comparison with existing bulk expression data based methods in breast cancer prognosis, our method shows a better performance. Our single-cell-based signatures provide clues to the relation between EMT and clinical outcomes of breast cancer. In addition, the proposed method can also be useful when applied to scRNA-seq datasets of different biological processes other than EMT.

MotivationUnravelling cancer driver genes is important in cancer research. Although computational methods have been developed to identify cancer drivers, most of them detect cancer drivers at population level. However, two patients who have the same cancer type and receive the same treatment may have different outcomes because each patient has a different genome and their disease might be driven by different driver genes. Therefore new methods are being developed for discovering cancer drivers at individual level, but existing personalised methods only focus on coding drivers while microRNAs (miRNAs) have been shown to drive cancer progression as well. Thus, novel methods are required to discover both coding and miRNA cancer drivers at individual level.ResultsWe propose the novel method, pDriver, to discover personalised cancer drivers. pDriver includes two stages: (1) Constructing gene networks for each cancer patient and (2) Discovering cancer drivers for each patient based on the constructed gene networks. To demonstrate the effectiveness of pDriver, we have applied it to five TCGA cancer datasets and compared it with the state-of-the-art methods. The result indicates that pDriver is more effective than other methods. Furthermore, pDriver can also detect miRNA cancer drivers and most of them have been confirmed to be associated with cancer by literature. We further analyse the predicted personalised drivers for breast cancer patients and the result shows that they are significantly enriched in many GO processes and KEGG pathways involved in breast cancer.Availability and implementationpDriver is available at https://github.com/pvvhoang/pDriverContactThuc.Le@unisa.edu.auSupplementary informationSupplementary data are available at Bioinformatics online.

PC algorithm (PC) – named after its authors, Peter and Clark – is an advanced constraint based method for learning causal structures. However, it is a time-consuming algorithm since the number of independence tests is exponential to the number of considered variables. Attempts to parallelise PC have been studied intensively, for example, by distributing the tests to all computing cores in a single computer. However, no effort has been made to speed up PC through parallelising the conditional independence tests into a cluster of computers. In this work, we propose MrPC, a robust and efficient PC algorithm, to accelerate PC to serve causal discovery in distributed systems. Alongside with MrPC, we also propose a novel manner to model non-linear causal relationships in gene regulatory data using kernel functions. We evaluate our method and its variants in the task of building gene regulatory networks. Experimental results on benchmark datasets show that the proposed MrPCgains up to seven times faster than sequential PC implementation. In addition, kernel functions outperform conventional linear causal modelling approach across different datasets. Refereed/Peer-reviewed

A valid machine model is predictive, but a predictive model may not be valid. The gap between these two can be larger than many practitioners may expect. Refereed/Peer-reviewed

Autism Spectrum Disorder (ASD) is defined as polygenetic developmental and neurobiological disorders that cover a variety of development delays in social interactions. In recent years, computational methods using gene expression data have been proved to be effective in predicting ASD at the early stage. Feature selection methods directly affect the prediction performance of the ASD prognosis methods. With the advances of computational methods and exploding of high-dimensional ASD gene expression data, there is a need to examine the performance of different computational techniques in predicting ASD. In this paper, we review and conduct a comparison study of 22 different feature selection methods for predicting ASD from gene expression data. The methods are categorised into traditional methods (14 methods) and network-based methods (8 methods). The experimental results have shown that the network-based methods generally outperform the traditional feature selection methods in all three accuracy measures, including AUC (area under the curve), F1-score, and Matthews Correlation Coefficient Refereed/Peer-reviewed

Causal feature selection has attracted much attention in recent years, as the causal features selected imply the causal mechanism related to the class attribute, leading to more reliable prediction models built using them. Currently there is a need of developing multi-source feature selection methods, since in many applications data for studying the same problem has been collected from various sources, such as multiple gene expression datasets obtained from different experiments for studying the causes of the same disease. However, the state-of-the-art causal feature selection methods generally tackle a single dataset, anda direct application of the methods to multiple datasets will result in unreliable results as the datasets may have different distributions. To address the challenges, by utilizing the concept of causal invariance in causal inference, we firstly formulate the problem of causal feature selection with multiple datasets as a search problem for an invariant set across the datasets, then give the upper and lower bounds of the invariant set, and finally we propose a new Multi-source Causal Feature Selection algorithm, MCFS. Using synthetic and real world datasets and 16 feature selection methods, the extensive experiments have validated the effectiveness of MCFS. Refereed/Peer-reviewed

Algorithmic fairness is an important aspect when data is used for predictive purposes. This paper analyses the sources of discrimination, and proposes sufficient conditions for building non-discriminatory/fair predictive models in the presence of context attributes. The paper then uses real world datasets to demonstrate the existence and the extent of discrimination in applications and the independence between the discrimination of datasets and the discrimination of classification models. Refereed/Peer-reviewed

With more and more decisions being made by learnt algorithms from data, algorithmic discriminations have become a risk for civil rights. The detection of discrimination is a process of counterfactual reasoning. This paper proposes a general detection framework by combining a data mining method with a well established counterfactual reasoning framework, potential outcome model. The potential outcome model supports operational definitions of global and local discriminations and discriminations by combined factors, while a data mining method makes the detection efficient. The proposed method, instantiated by association rule mining with potential outcome model based causal effect estimation, is evaluated with four real world data sets and is compared with a Bayesian network (BN) based detection method. It is able to detect not only global discriminations that are detected by the BN based method, but also local and combined discriminations that the BN based method cannot find. The proposed method is efficient, and scales well with the data set size and the number of attributes. Refereed/Peer-reviewed

After publication of this supplement article [1], it was brought to our attention that the Fig. 3 was incorrect. The correct Fig. 3 is as below

Single-cell RNA-seq (scRNAseq) technologies are rapidly evolving and a growing number of datasets are now available. While very informative, in standard scRNAseq experiments the spatial organization of the cells in the organism or tissue of origin is lost. Conversely, spatial RNA-seq technologies designed to keep the localization of the cells have limited throughput and gene coverage. Mapping scRNAseq to data of genes with spatial information can thus increase coverage while providing spatial location. However, methods to perform such a mapping are still in their infancy and have not been benchmarked in an unbiased manner. To bridge the gap, we organized the DREAM Single-Cell Transcriptomics challenge to evaluate methods for reconstructing the spatial arrangement of single cells from single-cell RNA sequencing data. The challenge focused on the spatial reconstruction of cells from the Drosophila embryo from single-cell transcriptomic and, leveraging as gold standard, in situ hybridization data of a set of selected driver genes from the Berkeley Drosophila Transcription Network Project reference atlas. The 34 participating teams used an array of different algorithms for gene selection and location prediction. We devised a novel scoring and cross-validation scheme to evaluate the robustness of the best performing algorithms. Participants were able to correctly and robustly localize rare subpopulations of cells, accurately mapping both spatially co-localized and scattered groups of cells. The selection of predictor genes was essential for accurately locating the cells in the embryo. Among the most frequently selected set of genes we measured a relatively high expression entropy, high spatial clustering and the presence of prominent developmental genes such as gap and pair-ruled genes and tissue defining markers.

Background The development of new drugs is costly, time consuming, and often accompanied with safety issues. Drug repurposing can avoid the expensive and lengthy process of drug development by finding new uses for already approved drugs. In order to repurpose drugs effectively, it is useful to know which proteins are targeted by which drugs. Computational models that estimate the interaction strength of new drug–target pairs have the potential to expedite drug repurposing. Several models have been proposed for this task. However, these models represent the drugs as strings, which are not a natural way to represent molecules. Methods We propose a new model called GraphDTA that represents drugs as graphs and uses graph neural networks to predict drug–target affinity. We test 4 graph neural network variants, including GCN, GAT, GIN, and a combined GAT-GCN architecture, for the task of drug–affinity prediction. We benchmark the performance of these models on the Davis and Kiba datasets. Results We show that graph neural networks not only predict drug–target affinity better than non-deep learning models, but also outperform competing deep learning methods. Of note, the GIN method performs consistently well for two separate benchmark datasets and for two key performance metrics. In a post-hoc analysis of our model, we find that a graph neural network can learn the importance of known molecular descriptors without any prior knowledge. We also examine the model’s performance and find that a handful of drugs contribute disproportionately to the total prediction error. Conclusions Our results confirm that deep learning models are appropriate for drug–target binding affinity prediction, and that representing drugs as graphs can lead to further improvements. Although we focus on drug–target affinity prediction, our GraphDTA model is a generic solution for any collaborating filtering or recommendation problem where either data input can be represented as a graph.

BackgroundA microRNA (miRNA) sponge is an RNA molecule with multiple tandem miRNA response elements that can sequester miRNAs from their target mRNAs. Despite growing appreciation of the importance of miRNA sponges, our knowledge of their complex functions remains limited. Moreover, there is still a lack of miRNA sponge research tools that help researchers to quickly compare their proposed methods with other methods, apply existing methods to new datasets, or select appropriate methods for assisting in subsequent experimental design.ResultsTo fill the gap, we present an R/Bioconductor package, miRsponge, for simplifying the procedure of identifying and analyzing miRNA sponge interaction networks and modules. It provides seven popular methods and an integrative method to identify miRNA sponge interactions. Moreover, it supports the validation of miRNA sponge interactions and the identification of miRNA sponge modules, as well as functional enrichment and survival analysis of miRNA sponge modules.ConclusionsThis package enables researchers to quickly evaluate their new methods, apply existing methods to new datasets, and consequently speed up miRNA sponge research.

The classification of clinical samples based on gene expression data is an important part of precision medicine. However, it has proved difficult to accurately predict survival outcomes and treatment responses for cancer patients. In this manuscript, we show how transforming gene expression data into a set of personalized (sample-specific) networks can allow us to harness existing graph-based methods to improve classifier performance. Existing approaches to personalized gene networks all have the limitation that they depend on other samples in the data and must get re-computed whenever a new sample is introduced. Here, we propose a novel method, called Personalized Annotation-based Networks (PAN), that avoids this limitation by using curated annotation databases to transform gene expression data into a graph. These databases organize genes into overlapping gene sets, called annotations, that we use to build a network where nodes represent functional terms and edges represent the similarity between them. Unlike competing methods, PANs are calculated for each sample independent of the population, making it a more efficient way to obtain single-sample networks. Using three breast cancer datasets as a case study (METABRIC and a super-set of GEO studies), we show that PAN classifiers not only predict cancer relapse better than gene features alone, but also outperform PPI and population-level graph-based classifiers. This work demonstrates the practical advantages of graph-based classification for high-dimensional genomic data, while offering a new approach to making sample-specific networks.Supplementary informationThe codes and data are available at https://github.com/thinng/PAN.ContactThuc.Le@unisa.edu.au

Outliers are objects that are significantly different from the others in the same dataset. They often contain insightful information for understanding the data and the data generation process. Traditional outlier detection methods can generally be divided into two categories: model-based and proximity-based approaches. A new type of model-based approach has recently been proposed to use the Bayesian network (BN) framework to discover more meaningful outliers with better interpretability. They yield very good detection result when anomalousness is mainly due to the violation of the dependency among variables. However, when anomalousness is caused by reasons other than dependency violation, BN-based methods produce very poor performance. To address this problem, we propose an ensemble outlier detection method that combines BN-based and proximity-based techniques to achieve more stable outlier detection results in different scenarios. To our best knowledge, the proposed method is the first to bring together the two major categories of outlier detection techniques. Comprehensive experiments have been done on both synthetic and real world datasets, and the results show that our method outperforms the baseline methods in most cases. Refereed/Peer-reviewed

Estimating heterogeneous treatment effects is an important problem in many medical and biological applications since treatments may have different effects on the prognoses of different patients. Recently, several recursive partitioning methods have been proposed to identify the subgroups that with different responds to a treatment, and they rely on a fitness criterion to minimize the error between the estimated treatment effects and the unobservable true effects. In this paper, we propose that a heterogeneity criterion, which maximizes the differences of treatment effects among the subgroups, also needs to be considered. Moreover, we show that better performances can be achieved when the fitness and the heterogeneous criteria are considered simultaneously. Selecting the optimal splitting points then becomes a multi-objective problem; however, a solution that achieves optimal in both aspects are often not available. To solve this problem, we propose a multi-objective splitting procedure to balance both criteria. The proposed procedure is computationally efficient and fits naturally into the existing recursive partitioning framework. Experimental results show that the proposed multi-objective approach performs consistently better than existing ones.Author summaryThe effects of a treatment are often not the same for different individuals with different gene expressions. Learning to predict the heterogeneous treatment effects from clinical and expression data is an important step towards personalized medical treatment. Existing computational methods are not ideal for the task because they do not address the interpretability of the model and do not consider the limited sample sizes in biological and medical applications. Our method addresses these issues and achieves superior performance in analyzing the treatment effects of radiotherapy on breast cancer patients.

With the increasing need of personalised decision making, such as personalised medicine and online recommendations, a growing attention has been paid to the discovery of the context and heterogeneity of causal relationships. Most existing methods, however, assume a known cause (e.g. a new drug) and focus on identifying from data the contexts of heterogeneous effects of the cause (e.g. patient groups with different responses to the new drug). There is no approach to efficiently detecting directly from observational data context specific causal relationships, i.e. discovering the causes and their contexts simultaneously. In this paper, by taking the advantages of highly efficient decision tree induction and the well established causal inference framework, we propose the Tree based Context Causal rule discovery (TCC) method, for efficient exploration of context specific causal relationships from data. Experiments with both synthetic and real world data sets show that TCC can effectively discover context specific causal rules from the data., Comment: This paper has been accepted by Intelligent Data Analysis

Discovering causal relationships from genomic data is the ultimate goal in gene regulation research. Constraint based causal exploration algorithms, such as PC, FCI, RFCI, PC-simple, IDA and Joint-IDA have achieved significant progress and have many applications. However, their applications in bioinformatics are still limited due to their high computational complexity. In this paper, we present an R package, ParallelPC, that includes the parallelised versions of these causal exploration algorithms and 12 different conditional independence tests for each. The parallelised algorithms help speed up the procedure of experimenting large biological datasets and reduce the memory used when running the algorithms. Our experiment results on a real gene expression dataset show that using the parallelised algorithms it is now practical to explore causal relationships in high dimensional datasets with thousands of variables in a personal multicore computer. We present some typical applications in bioinformatics using different algorithms in ParallelPC. ParallelPC is available in CRAN repository at https://cran.r-project.org/web/packages/ParallelPC/index.html Refereed/Peer-reviewed

Cross-view data are collected from two different views or sources about the same subjects. As the information from these views often consolidate and/or complement each other, cross-view data analysis can gain more insights for decision making. A main challenge of cross-view data analysis is how to effectively explore the inherently correlated and high-dimensional data. Dimension reduction offers an effective solution for this problem. However, how to choose right models and parameters involved for dimension reduction is still an open problem. In this paper, we propose an effective sparse learning algorithm for cross-view dimensionality reduction. A distinguished character of our model selection is that it is nonparametric and automatic. Specifically, we represent the correlation of cross-view data using a covariance matrix. Then, we decompose the matrix into a sequence of low-rank ones by solving an optimization problem in an alternating least squares manner. More importantly, a new and nonparametric sparsity-inducing function is developed to derive a parsimonious model. Extensive experiments are conducted on real-world data sets to evaluate the effectiveness of the proposed algorithm. The results show that our method is competitive with the state-of-the-art sparse learning algorithms. Refereed/Peer-reviewed

In recent years, many methods have been developed for detecting causal relationships in observational data. Some of them have the potential to tackle large data sets. However, these methods fail to discover a combined cause, i.e. a multi-factor cause consisting of two or more component variables which individually are not causes. A straightforward approach to uncovering a combined cause is to include both individual and combined variables in the causal discovery using existing methods, but this scheme is computationally infeasible due to the huge number of combined variables. In this paper, we propose a novel approach to address this practical causal discovery problem, i.e. mining combined causes in large data sets. The experiments with both synthetic and real world data sets show that the proposed method can obtain high-quality causal discoveries with a high computational efficiency., This paper has been accepted by Knowledge-Based Systems

Uncovering causal relationships in data is a major objective of data analytics. Currently, there is a need for scalable and automated methods for causal relationship exploration in data. Classification methods are fast and they could be practical substitutes for finding causal signals in data. However, classification methods are not designed for causal discovery and a classification method may find false causal signals and miss the true ones. In this paper, we develop a causal decision tree (CDT) where nodes have causal interpretations. Our method follows a well-established causal inference framework and makes use of a classic statistical test to establish the causal relationship between a predictor variable and the outcome variable. At the same time, by taking the advantages of normal decision trees, a CDT provides a compact graphical representation of the causal relationships, and the construction of a CDT is fast as a result of the divide and conquer strategy employed, making CDTs practical for representing and finding causal signals in large data sets. Experiment results demonstrate that CDTs can identify meaningful causal relationships and the CDT algorithm is scalable. Refereed/Peer-reviewed

Partial association test is a statistical means to test whether the association between two variables is persistent given other variables. The CR-PA algorithm has been developed to identify causal rules (CRs) by integrating association rule mining and partial association (PA) tests. The use of association rule mining enables fast identification of causal hypotheses (association rules) from large data sets, and partial association tests on these association rules eliminate non-persistent associations. This chapter firstly describes the basics of partial association tests and association rule mining, and then the CR-PA algorithm is presented in detail, followed by the discussions on the complexity and false discoveries of the algorithm. A tool which implements CR-PA is also introduced.

In this chapter, we present CR-CS, a different method for mining causal rules. Similar to CR-PA introduced in the previous chapter, CR-CS uses association rule mining to generate causal hypotheses, but for validating the hypotheses, CR-CS utilises the idea of a retrospective cohort study (CS). For each candidate causal rule, CR-CS firstly creates a fair data set which consists of pair-wise matched samples in the original data set and then it tests the association of the LHS (predictor) and the RHS (target) of the rule based on the fair data set. The creation of a fair data set imitates the process of collecting samples in a cohort study, for which each sample in the exposure group (LHS = 1) has a matching sample in the control group (LHS = 0) with respect to the values of control variables. Therefore an association confirmed using a fair data set is expected to be a persistent association across different values of the control variables, since matching has been employed to eliminate the effect of control variables. In this chapter, after introducing the motivation and general idea of CR-CS, the test of associations used by CR-CS is described. Then the CR-CS algorithm is presented in detail, followed by the discussions on the algorithm. Finally a tool which implements CR-CS is illustrated with examples.

In Chaps. 2–5 we have presented four methods for discovering the causal relationships around a given target variable, focusing on the basic ideas and procedures of the methods. In this chapter, we apply the methods to a series of data sets to evaluate and compare their efficiency. Discussions on the experimental results are provided in this chapter too.

Randomised controlled trials (RCTs) are the most effective approach to causal discovery, but in many circumstances it is impossible to conduct RCTs. Therefore observational studies based on passively observed data are widely accepted as an alternative to RCTs. However, in observational studies, prior knowledge is required to generate the hypotheses about the cause-effect relationships to be tested, hence they can only be applied to problems with available domain knowledge and a handful of variables. In practice, many data sets are of high dimensionality, which leaves observational studies out of the opportunities for causal discovery from such a wealth of data sources. In another direction, many efficient data mining methods have been developed to identify associations among variables in large data sets. The problem is, causal relationships imply associations, but the reverse is not always true. However we can see the synergy between the two paradigms here. Specifically, association rule mining can be used to deal with the high-dimensionality problem while observational studies can be utilised to eliminate non-causal associations. In this paper we propose the concept of causal rules (CRs) and develop an algorithm for mining CRs in large data sets. We use the idea of retrospective cohort studies to detect CRs based on the results of association rule mining. Experiments with both synthetic and real world data sets have demonstrated the effectiveness and efficiency of CR mining. In comparison with the commonly used causal discovery methods, the proposed approach in general is faster and has better or competitive performance in finding correct or sensible causes. It is also capable of finding a cause consisting of multiple variables, a feature that other causal discovery methods do not possess., Comment: This paper has been accepted by ACM TIST journal and will be available soon

This chapter presents the PC-simple algorithm and illustrates how to use the algorithm in the exploration for local causal relationships around a target variable. PC-simple is a simplified version of the PC algorithm, a classic method for learning a complete casual Bayesian network. We firstly discuss how the PC algorithm establishes causal relationships by the way of detecting persistent associations, then we introduce PC-simple in detail, followed by the discussions on PC-simple. The last section of this chapter introduces the R implementation of PC-simple.

Discovering causal relationships is the ultimate goal of many scientific explorations. Causal relationships can be identified with controlled experiments, but such experiments are often very expensive and sometimes impossible to conduct. On the other hand, the collection of observational data has increased dramatically in recent decades. Therefore it is desirable to find causal relationships from the data directly. Significant progress has been made in the field of discovering causal relationships using the Causal Bayesian Network (CBN) theory. The applications of CBNs, however, are greatly limited due to the high computational complexity. In another direction, association rule mining has been shown to be an efficient data mining means for relationship discovery. However, although causal relationships imply associations, the reverse does not always hold. In this paper we study how to use an efficient association mining approach to discover potential causal rules in observational data.We make use of the idea of retrospective cohort studies, a widely used approach in medical and social research, to detect causal association rules. In comparison with the constraint-based methods within the CBN paradigm, the proposed approach is faster and is capable of finding a cause consisting of combined variables. Refereed/Peer-reviewed

As a basic and effective tool for explanation, prediction and decision making, causal relationships have been utilized in almost all disciplines. Traditionally, causal relationships are identified by making use of interventions or randomized controlled experiments. However, conducting such experiments is often expensive or even impossible due to cost or ethical concerns. Therefore, there has been an increasing interest in discovering causal relationships based on observational data, and in the past few decades, significant contributions have been made to this field by computer scientists., Inspired by such achievements and following the success of CD 2016 - CD 2020, CD 2021 continues to serve as a forum for researchers and practitioners in data mining and other disciplines to share their recent research in causal discovery in their respective fields and to explore the possibility of interdisciplinary collaborations in the study of causality. Based on the platform of KDD, this workshop is especially interested in attracting contributions that link data mining/machine learning research with causal discovery, and solutions to causal discovery in large scale datasets.