Osteosarcoma (OS) is the most common primary malignant bone tumor of both children and pet canines. Its characteristic genomic instability and complexity coupled with the dearth of knowledge about its etiology has made improvement in the current treatment difficult. We use the existing literature about the biological pathways active in OS and combine it with the current research involving natural compounds to identify new targets and design more effective drug therapies. The key components of these pathways are modeled as a Boolean network with multiple inputs and multiple outputs. The combinatorial circuit is employed to theoretically predict the efficacies of various drugs in combination with Cryptotanshinone. We show that the action of the herbal drug, Cryptotanshinone on OS cell lines induces apoptosis by increasing sensitivity to TNF-related apoptosis-inducing ligand (TRAIL) through its multi-pronged action on STAT3, DRP1 and DR5. The Boolean framework is used to detect additional drug intervention points in the pathway that could amplify the action of Cryptotanshinone.

In this work, we develop a systematic approach for applying pathway knowledge to a multivariate Gaussian mixture model for dissecting a heterogeneous cancer tissue. The downstream transcription factors are selected as observables from available partial pathway knowledge in such a way that the subpopulations produce some differential behavior in response to the drugs selected in the upstream. For each subpopulation, each unique (drug, observable) pair is considered as a unique dimension of a multivariate Gaussian distribution. Expectation-maximization (EM) algorithm with hill-climbing is then used to rank the most probable estimates of the mixture composition based on the log-likelihood value. A major contribution of this work is to examine the efficacy of the EM based approach in estimating the composition of experimental mixture sets from cell-by-cell measurements collected on a dynamic cell imaging platform. Towards this end, we apply the algorithm on hourly data collected for two different mixture compositions of A2058, HCT116, and SW480 cell lines for three scenarios: untreated, Lapatinib-treated, and Temsirolimus-treated. Additionally, we show how this methodology can provide a basis for comparing the killing rate of different drugs for a heterogeneous cancer tissue. This obviously has important implications for designing efficient drugs for treating heterogeneous malignant tumors.

The number of deaths associated with Pancreatic Cancer has been on the rise in the United States making it an especially dreaded disease. The overall prognosis for pancreatic cancer patients continues to be grim because of the complexity of the disease at the molecular level involving the potential activation/inactivation of several diverse signaling pathways. In this paper, we first model the aberrant signaling in pancreatic cancer using a multi-fault Boolean Network. Thereafter, we theoretically evaluate the efficacy of different drug combinations by simulating this boolean network with drugs at the relevant intervention points and arrive at the most effective drug(s) to achieve cell death. The simulation results indicate that drug combinations containing Cryptotanshinone, a traditional Chinese herb derivative, result in considerably enhanced cell death. These in silico results are validated using wet lab experiments we carried out on Human Pancreatic Cancer (HPAC) cell lines.

Signaling pathways oversee highly efficient cellular mechanisms such as growth, division, and death. These processes are controlled by robust negative feedback loops that inhibit receptor-mediated growth factor pathways. Specifically, the ERK, the AKT, and the S6K feedback loops attenuate signaling via growth factor receptors and other kinase receptors to regulate cell growth. Irregularity in any of these supervised processes can lead to uncontrolled cell proliferation and possibly Cancer. These irregularities primarily occur as mutated genes, and an exhaustive search of the perfect drug combination by performing experiments can be both costly and complex. Hence, in this paper, we model the Lung Cancer pathway as a Modified Boolean Network that incorporates feedback. By simulating this network, we theoretically predict the drug combinations that achieve the desired goal for the majority of mutations. Our theoretical analysis identifies Cryptotanshinone, a traditional Chinese herb derivative, as a potent drug component in the fight against cancer. We validated these theoretical results using multiple wet lab experiments carried out on H2073 and SW900 lung cancer cell lines.

BACKGROUND: Cancer Tissue Heterogeneity is an important consideration in cancer research as it can give insights into the causes and progression of cancer. It is known to play a significant role in cancer cell survival, growth and metastasis. Determining the compositional breakup of a heterogeneous cancer tissue can also help address the therapeutic challenges posed by heterogeneity. This necessitates a low cost, scalable algorithm to address the challenge of accurate estimation of the composition of a heterogeneous cancer tissue. METHODS: In this paper, we propose an algorithm to tackle this problem by utilizing the data of accurate, but high cost, single cell line cell-by-cell observation methods in low cost aggregate observation method for heterogeneous cancer cell mixtures to obtain their composition in a Bayesian framework. RESULTS: The algorithm is analyzed and validated using synthetic data and experimental data. The experimental data is obtained from mixtures of three separate human cancer cell lines, HCT116 (Colorectal carcinoma), A2058 (Melanoma) and SW480 (Colorectal carcinoma). CONCLUSION: The algorithm provides a low cost framework to determine the composition of heterogeneous cancer tissue which is a crucial aspect in cancer research.

Tumor heterogeneity has been increasingly recognized as one of the potentially contributing factors in explaining drug resistance. In order to gain better understanding of heterogeneous cancer cell populations and different cells' responses to various drugs, we use fluorescent proteins to mark the cells and a live-cell dynamic imaging platform to collect cell-by-cell measurements. After addressing the issue of fluorescent reporter variance in a Bayesian inference framework, we decompose the different cell types in the mixture and calculate their proportions and counts over time responding to different drug treatments. Additionally, the drug response as characterized by the cell death rate was also computed for these cells, and their different sensitivity was demonstrated. Overall, this work represents an important advancement toward evaluating cancer heterogeneity and drug responses in heterogeneous cancer cell populations.

MOTIVATION: In small-sample settings, bolstered error estimation has been shown to perform better than cross-validation and competitively with bootstrap with regard to various criteria. The key issue for bolstering performance is the variance setting for the bolstering kernel. Heretofore, this variance has been determined in a non-parametric manner from the data. Although bolstering based on this variance setting works well for small feature sets, results can deteriorate for high-dimensional feature spaces. RESULTS: This article computes an optimal kernel variance depending on the classification rule, sample size, model and feature space, both the original number and the number remaining after feature selection. A key point is that the optimal variance is robust relative to the model. This allows us to develop a method for selecting a suitable variance to use in real-world applications where the model is not known, but the other factors in determining the optimal kernel are known. AVAILABILITY: Companion website at http://compbio.tgen.org/paper_supp/high_dim_bolstering. CONTACT: [email protected].

MOTIVATION: The receiver operator characteristic (ROC) curves are commonly used in biomedical applications to judge the performance of a discriminant across varying decision thresholds. The estimated ROC curve depends on the true positive rate (TPR) and false positive rate (FPR), with the key metric being the area under the curve (AUC). With small samples these rates need to be estimated from the training data, so a natural question arises: How well do the estimates of the AUC, TPR and FPR compare with the true metrics? RESULTS: Through a simulation study using data models and analysis of real microarray data, we show that (i) for small samples the root mean square differences of the estimated and true metrics are considerable; (ii) even for large samples, there is only weak correlation between the true and estimated metrics; and (iii) generally, there is weak regression of the true metric on the estimated metric. For classification rules, we consider linear discriminant analysis, linear support vector machine (SVM) and radial basis function SVM. For error estimation, we consider resubstitution, three kinds of cross-validation and bootstrap. Using resampling, we show the unreliability of some published ROC results. AVAILABILITY: Companion web site at http://compbio.tgen.org/paper_supp/ROC/roc.html CONTACT: [email protected].

MOTIVATION: It is commonplace for authors to propose a new classification rule, either the operator construction part or feature selection, and demonstrate its performance on real data sets, which often come from high-dimensional studies, such as from gene-expression microarrays, with small samples. Owing to the variability in feature selection and error estimation, individual reported performances are highly imprecise. Hence, if only the best test results are reported, then these will be biased relative to the overall performance of the proposed procedure. RESULTS: This article characterizes reporting bias with several statistics and computes these statistics in a large simulation study using both modeled and real data. The results appear as curves giving the different reporting biases as functions of the number of samples tested when reporting only the best or second best performance. It does this for two classification rules, linear discriminant analysis (LDA) and 3-nearest-neighbor (3NN), and for filter and wrapper feature selection, t-test and sequential forward search. These were chosen on account of their well-studied properties and because they were amenable to the extremely large amount of processing required for the simulations. The results across all the experiments are consistent: there is generally large bias overriding what would be considered a significant performance differential, when reporting the best or second best performing data set. We conclude that there needs to be a database of data sets and that, for those studies depending on real data, results should be reported for all data sets in the database. AVAILABILITY: Companion web site at http://gsp.tamu.edu/Publications/supplementary/yousefi09a/

MOTIVATION: High-throughput technologies for rapid measurement of vast numbers of biological variables offer the potential for highly discriminatory diagnosis and prognosis; however, high dimensionality together with small samples creates the need for feature selection, while at the same time making feature-selection algorithms less reliable. Feature selection must typically be carried out from among thousands of gene-expression features and in the context of a small sample (small number of microarrays). Two basic questions arise: (1) Can one expect feature selection to yield a feature set whose error is close to that of an optimal feature set? (2) If a good feature set is not found, should it be expected that good feature sets do not exist? RESULTS: The two questions translate quantitatively into questions concerning conditional expectation. (1) Given the error of an optimal feature set, what is the conditionally expected error of the selected feature set? (2) Given the error of the selected feature set, what is the conditionally expected error of the optimal feature set? We address these questions using three classification rules (linear discriminant analysis, linear support vector machine and k-nearest-neighbor classification) and feature selection via sequential floating forward search and the t-test. We consider three feature-label models and patient data from a study concerning survival prognosis for breast cancer. With regard to the two focus questions, there is similarity across all experiments: (1) One cannot expect to find a feature set whose error is close to optimal, and (2) the inability to find a good feature set should not lead to the conclusion that good feature sets do not exist. In practice, the latter conclusion may be more immediately relevant, since when faced with the common occurrence that a feature set discovered from the data does not give satisfactory results, the experimenter can draw no conclusions regarding the existence or nonexistence of suitable feature sets. AVAILABILITY: http://ee.tamu.edu/~edward/feature_regression/

Abstract Motivation: Ranking feature sets is a key issue for classification, for instance, phenotype classification based on gene expression. Since ranking is often based on error estimation, and error estimators suffer to differing degrees of imprecision in small-sample settings, it is important to choose a computationally feasible error estimator that yields good feature-set ranking. Results: This paper examines the feature-ranking performance of several kinds of error estimators: resubstitution, cross-validation, bootstrap and bolstered error estimation. It does so for three classification rules: linear discriminant analysis, three-nearest-neighbor classification and classification trees. Two measures of performance are considered. One counts the number of the truly best feature sets appearing among the best feature sets discovered by the error estimator and the other computes the mean absolute error between the top ranks of the truly best feature sets and their ranks as given by the error estimator. Our results indicate that bolstering is superior to bootstrap, and bootstrap is better than cross-validation, for discovering top-performing feature sets for classification when using small samples. A key issue is that bolstered error estimation is tens of times faster than bootstrap, and faster than cross-validation, and is therefore feasible for feature-set ranking when the number of feature sets is extremely large. Availability: We provide a companion website, which contains the complete set of tables and plots regarding the simulation study, and a compilation of references on feature-set ranking with applications in Genomics. The companion website can be accessed at the URL http://ee.tamu.edu/~edward/bolster_ranking Contact: [email protected]