Generative latent variable models (GLVMs) are prevalent in developing soft sensors for predicting industrial process quality indices owing to their exceptional capability in extracting features, reducing data dimensionality, and explaining data generation mechanisms. Dynamical GLVMs (DGLVMs) are able to effectively deal with the dynamical characteristics of industrial data changes over time, exhibiting prominence in time-series prediction and dynamical process modeling. However, the existing DGLVM-based soft sensors share an intrinsic imperfection that they focus on capturing the temporal correlations between variables while disregarding the causal relationships between variables. Based on industrial data, it is difficult to recover the causalities between variables from the correlations, resulting in compromised generalization performance of the existing DGLVM-based soft sensors. In light of this limitation, a novel causal knowledge-assisted semi-supervised DGLVM (CK-SsDGLVM) is proposed, and an ad hoc parameter learning method based on the expectation–maximization (EM) algorithm is developed to train the CK-SsDGLVM. The performance of the CK-SsDGLVM is comprehensively evaluated using an artificial numerical case and an actual industrial process. The experimental results demonstrate that the CK-SsDGLVM could achieve superior generalization accuracy and interpretability compared to benchmark models.

DNA copy number variations (CNVs) are prevalent in all types of tumors. It is still a challenge to study how CNVs play a role in driving tumorgenic mechanisms that are either universal or specific in different cancer types. To address the problem, we introduce a transfer learning framework to discover common CNVs shared across different tumor types as well as CNVs specific to each tumor type from genome-wide CNV data measured by array CGH and SNP genotyping array. The proposed model, namely Transfer Learning with Fused LASSO (TLFL), detects latent CNV components from multiple CNV datasets of different tumor types to distinguish the CNVs that are common across the datasets and those that are specific in each dataset. Both the common and type-specific CNVs are detected as latent components in matrix factorization coupled with fused LASSO on adjacent CNV probe features. TLFL considers the common latent components underlying the multiple datasets to transfer knowledge across different tumor types. In simulations and experiments on real cancer CNV datasets, TLFL detected better latent components that can be used as features to improve classification of patient samples in each individual dataset compared with the model without the knowledge transfer. In cross-dataset analysis on bladder cancer and cross-domain analysis on breast cancer and ovarian cancer, TLFL also learned latent CNV components that are both predictive of tumor stages and correlate with known cancer genes.

Detecting DNA copy number variations (CNVs) from arrayCGH or genotyping-array data to correlate with cancer outcomes is crucial for understanding the molecular mechanisms underlying cancer. Previous methods either focus on detecting CNVs in each individual patient sample or common CNVs across all the patient samples. These methods ignore the discrepancies introduced by the heterogeneity in the patient samples, which implies that common CNVs might only be shared within some groups of samples instead of all samples. In this paper, we propose a latent feature model that couples sparse sample group selection with fused lasso on CNV components to identify group-specific CNVs. Assuming a given group structure on patient samples by clinical information, sparse group selection on fused lasso (SGS-FL) identifies the optimal latent CNV components, each of which is specific to the samples in one or several groups. The group selection for each CNV component is determined dynamically by an adaptive algorithm to achieve a desired sparsity. Simulation results show that SGS-FL can more accurately identify the latent CNV components when there is a reliable underlying group structure in the samples. In the experiments on arrayCGH breast cancer and bladder cancer datasets, SGS-FL detected CNV regions that are more relevant to cancer, and provided latent feature weights that can be used for better sample classification.

Graph-based semi-supervised learning improves classification by combining labeled and unlabeled data through label propagation. It was shown that the sparse representation of graph by weighted local neighbors provides a better similarity measure between data points for label propagation. However, selecting local neighbors can lead to disjoint components and incorrect neighbors in graph, and thus, fail to capture the underlying global structure. In this paper, we propose to learn a nonnegative low-rank graph to capture global linear neighborhoods, under the assumption that each data point can be linearly reconstructed from weighted combinations of its direct neighbors and reachable indirect neighbors. The global linear neighborhoods utilize information from both direct and indirect neighbors to preserve the global cluster structures, while the low-rank property retains a compressed representation of the graph. An efficient algorithm based on a multiplicative update rule is designed to learn a nonnegative low-rank factorization matrix minimizing the neighborhood reconstruction error. Large scale simulations and experiments on UCI datasets and high-dimensional gene expression datasets showed that label propagation based on global linear neighborhoods captures the global cluster structures better and achieved more accurate classification results.

MOTIVATION: Array comparative genomic hybridization (arrayCGH) is widely used to measure DNA copy numbers in cancer research. ArrayCGH data report log-ratio intensities of thousands of probes sampled along the chromosomes. Typically, the choices of the locations and the lengths of the probes vary in different experiments. This discrepancy in choosing probes poses a challenge in integrated classification or analysis across multiple arrayCGH datasets. We propose an alignment-based framework to integrate arrayCGH samples generated from different probe sets. The alignment framework seeks an optimal alignment between the probe series of one arrayCGH sample and the probe series of another sample, intended to find the maximum possible overlap of DNA copy number variations between the two measured chromosomes. An alignment kernel is introduced for integrative patient sample classification and a multiple alignment algorithm is also introduced for identifying common regions with copy number aberrations. RESULTS: The probe alignment kernel and the MPA algorithm were experimented to integrate three bladder cancer datasets as well as artificial datasets. In the experiments, by integrating arrayCGH samples from multiple datasets, the probe alignment kernel used with support vector machines significantly improved patient sample classification accuracy over other baseline kernels. The experiments also demonstrated that the multiple probe alignment (MPA) algorithm can find common DNA aberrations that cannot be identified with the standard interpolation method. Furthermore, the MPA algorithm also identified many known bladder cancer DNA aberrations containing four known bladder cancer genes, three of which cannot be detected by interpolation. AVAILABILITY: http://www.cs.umn.edu/compbio/ProbeAlign.

MOTIVATION: Incorporating biological prior knowledge into predictive models is a challenging data integration problem in analyzing high-dimensional genomic data. We introduce a hypergraph-based semi-supervised learning algorithm called HyperPrior to classify gene expression and array-based comparative genomic hybridization (arrayCGH) data using biological knowledge as constraints on graph-based learning. HyperPrior is a robust two-step iterative method that alternatively finds the optimal labeling of the samples and the optimal weighting of the features, guided by constraints encoding prior knowledge. The prior knowledge for analyzing gene expression data is that cancer-related genes tend to interact with each other in a protein-protein interaction network. Similarly, the prior knowledge for analyzing arrayCGH data is that probes that are spatially nearby in their layout along the chromosomes tend to be involved in the same amplification or deletion event. Based on the prior knowledge, HyperPrior imposes a consistent weighting of the correlated genomic features in graph-based learning. RESULTS: We applied HyperPrior to test two arrayCGH datasets and two gene expression datasets for both cancer classification and biomarker identification. On all the datasets, HyperPrior achieved competitive classification performance, compared with SVMs and the other baselines utilizing the same prior knowledge. HyperPrior also identified several discriminative regions on chromosomes and discriminative subnetworks in the PPI, both of which contain cancer-related genomic elements. Our results suggest that HyperPrior is promising in utilizing biological prior knowledge to achieve better classification performance and more biologically interpretable findings in gene expression and arrayCGH data. AVAILABILITY: http://compbio.cs.umn.edu/HyperPrior CONTACT: [email protected] SUPPLEMENTARY INFORMATION: Supplementary data are available at bioinformatics online.

Building reliable predictive models from multiple complementary genomic data for cancer study is a crucial step towards successful cancer treatment and a full understanding of the underlying biological principles. To tackle this challenging data integration problem, we propose a hypergraph-based learning algorithm called HyperGene to integrate microarray gene expressions and protein-protein interactions for cancer outcome prediction and biomarker identification. HyperGene is a robust two-step iterative method that alternatively finds the optimal outcome prediction and the optimal weighting of the marker genes guided by a protein-protein interaction network. Under the hypothesis that cancer-related genes tend to interact with each other, the HyperGene algorithm uses a protein-protein interaction network as prior knowledge by imposing a consistent weighting of interacting genes. Our experimental results on two large-scale breast cancer gene expression datasets show that HyperGene utilizing a curated protein-protein interaction network achieves significantly improved cancer outcome prediction. Moreover, HyperGene can also retrieve many known cancer genes as highly weighted marker genes.

MOTIVATION: A central problem in biomarker discovery from large-scale gene expression or single nucleotide polymorphism (SNP) data is the computational challenge of taking into account the dependence among all the features. Methods that ignore the dependence usually identify non-reproducible biomarkers across independent datasets. We introduce a new graph-based semi-supervised feature classification algorithm to identify discriminative disease markers by learning on bipartite graphs. Our algorithm directly classifies the feature nodes in a bipartite graph as positive, negative or neutral with network propagation to capture the dependence among both samples and features (clinical and genetic variables) by exploring bi-cluster structures in a graph. Two features of our algorithm are: (1) our algorithm can find a global optimal labeling to capture the dependence among all the features and thus, generates highly reproducible results across independent microarray or other high-thoughput datasets, (2) our algorithm is capable of handling hundreds of thousands of features and thus, is particularly useful for biomarker identification from high-throughput gene expression and SNP data. In addition, although designed for classifying features, our algorithm can also simultaneously classify test samples for disease prognosis/diagnosis. RESULTS: We applied the network propagation algorithm to study three large-scale breast cancer datasets. Our algorithm achieved competitive classification performance compared with SVMs and other baseline methods, and identified several markers with clinical or biological relevance with the disease. More importantly, our algorithm also identified highly reproducible marker genes and enriched functions from the independent datasets. AVAILABILITY: Supplementary results and source code are available at http://compbio.cs.umn.edu/Feature_Class. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.