Biomedical subjects

Qian Xie

Publications and source records attributed to Qian Xie.

28 records · Page 2Linked to original sources

PubMed2004

Multiclass Decision Forest--a novel pattern recognition method for multiclass classification in microarray data analysis.

The wealth of knowledge imbedded in gene expression data from DNA microarrays portends rapid advances in both research and clinic. Turning the prodigious and noisy data into knowledge is a challenge to the field of bioinformatics, and development of classifiers using supervised learning techniques is the primary methodological approach for clinical application using gene expression data. In this paper, we present a novel classification method, multiclass Decision Forest (DF), that is the direct extension of the two-class DF previously developed in our lab. Central to DF is the synergistic combining of multiple heterogenic but comparable decision trees to reach a more accurate and robust classification model. The computationally inexpensive multiclass DF algorithm integrates gene selection and model development, and thus eliminates the bias of gene preselection in crossvalidation. Importantly, the method provides several statistical means for assessment of prediction accuracy, prediction confidence, and diagnostic capability. We demonstrate the method by application to gene expression data for 83 small round blue-cell tumors (SRBCTs) samples belonging to one of four different classes. Based on 500 runs of 10-fold crossvalidation, tumor prediction accuracy was approximately 97%, sensitivity was approximately 95%, diagnostic sensitivity was approximately 91%, and diagnostic accuracy was approximately 99.5%. Among 25 genes selected to distinguish tumor class, 12 have functional information in the literature implicating their involvement in cancer. The four types of SRBCTs samples are also distinguishable in a clustering analysis based on the expression profiles of these 25 genes. The results demonstrated that the multiclass DF is an effective classification method for analysis of gene expression data for the purpose of molecular diagnostics.

Carcinoma, Small Cell↗

PubMed2004

Three new consensus QSAR models for the prediction of Ames genotoxicity.

Three QSAR methods, artificial neural net (ANN), k-nearest neighbors (kNN), and Decision Forest (DF), were applied to 3363 diverse compounds tested for their Ames genotoxicity. The ratio of mutagens to non-mutagens was 60/40 for this dataset. This group of compounds includes >300 therapeutic drugs. All models were developed using the same initial set of 148 topological indices: molecular connectivity chi indices and electrotopological state indices (atom-type, bond-type and group-type E-state), as well as binary indicators. While previous studies have found logP to be a determining factor in genotoxicity, it was not found to be important by any modeling method employed in this study. The three models yielded an average training/test concordance value of 88%, with a low percentage of false positives and false negatives. External validation testing on 400 compounds not used for QSAR model development gave an average concordance of 82%. This value increased to 92% upon removal of less reliable outcomes, as determined by a reliability criterion used within each model. The ANN model showed the best performance in predicting drug compounds, yielding 97% concordance (34/35 drugs) after the removal of less reliable predictions. The appreciable commonality found among the top 10 ranked descriptors from each model is of particular interest because of the diversity in the learning algorithms and descriptor selection techniques employed in this study. Forty percent of the most important descriptors in any one model are found in one or two other models. Fourteen of the most important descriptors relate directly to known toxicophores involved in potent genotoxic responses in Salmonella typhimurium. A comparison of the validation results with those of MULTICASE and DEREK indicated that the new models presented in this work perform substantially better than the former models in predicting genotoxicity of therapeutic drugs. Substantially higher specificity was achieved with these new models as compared with MULTICASE or DEREK with comparable sensitivities among all models.

Algorithms↗

PubMed2004

Using decision forest to classify prostate cancer samples on the basis of SELDI-TOF MS data: assessing chance correlation and prediction confidence.

Class prediction using "omics" data is playing an increasing role in toxicogenomics, diagnosis/prognosis, and risk assessment. These data are usually noisy and represented by relatively few samples and a very large number of predictor variables (e.g., genes of DNA microarray data or m/z peaks of mass spectrometry data). These characteristics manifest the importance of assessing potential random correlation and overfitting of noise for a classification model based on omics data. We present a novel classification method, decision forest (DF), for class prediction using omics data. DF combines the results of multiple heterogeneous but comparable decision tree (DT) models to produce a consensus prediction. The method is less prone to overfitting of noise and chance correlation. A DF model was developed to predict presence of prostate cancer using a proteomic data set generated from surface-enhanced laser deposition/ionization time-of-flight mass spectrometry (SELDI-TOF MS). The degree of chance correlation and prediction confidence of the model was rigorously assessed by extensive cross-validation and randomization testing. Comparison of model prediction with imposed random correlation demonstrated biologic relevance of the model and the reduction of overfitting in DF. Furthermore, two confidence levels (high and low confidences) were assigned to each prediction, where most misclassifications were associated with the low-confidence region. For the high-confidence prediction, the model achieved 99.2% sensitivity and 98.2% specificity. The model also identified a list of significant peaks that could be useful for biomarker identification. DF should be equally applicable to other omics data such as gene expression data or metabolomic data. The DF algorithm is available upon request.

Decision Support Techniques↗

PubMed2004

Assessment of prediction confidence and domain extrapolation of two structure-activity relationship models for predicting estrogen receptor binding activity.

Quantitative structure-activity relationship (QSAR) methods have been widely applied in drug discovery, lead optimization, toxicity prediction, and regulatory decisions. Despite major advances in algorithms and software, QSAR models have inherent limitations associated with a size and chemical-structure diversity of the training set, experimental error, and many characteristics of structure representation and correlation algorithms. Whereas excellent fit to the training data may be readily attainable, often models fail to predict accurately chemicals that are outside their domain of applicability. A QSAR's utility and, in the case of regulatory decisions, justification for usage increasingly depend on the ability to quantify a model's potential for predicting unknown chemicals with some known degree of certainty. It is never possible to predict an unknown chemical with absolute certainty. Here we report on two QSAR models based on different data sets for classification of chemicals according to their ability to bind to the estrogen receptor. The models were developed by using a novel QSAR method, Decision Forest, which combines the results of multiple heterogeneous but comparable Decision Tree models to produce a consensus prediction. We used an extensive cross-validation process to define an applicability domain for model predictions based on two quantitative measures: prediction confidence and domain extrapolation. Together, these measures quantify the accuracy of each prediction within and outside of the training domain. Despite being based on large and diverse training sets, both QSAR models had poor accuracy for chemicals within the domain of low confidence, whereas good accuracy was obtained for those within the domain of high confidence. For prediction in the high confidence domain, accuracy was inversely proportional to the degree of domain extrapolation. The model with a larger training set of 1,092, compared with 232 for the other, was more accurate in predicting chemicals at larger domain extrapolation, and could be particularly useful for rapidly prioritizing potential endocrine disruptors from large chemical universe.

Animals↗

PubMed2003

Study of 202 natural, synthetic, and environmental chemicals for binding to the androgen receptor.

A number of environmental and industrial chemicals are reported to possess androgenic or antiandrogenic activities. These androgenic endocrine disrupting chemicals may disrupt the endocrine system of humans and wildlife by mimicking or antagonizing the functions of natural hormones. The present study developed a low cost recombinant androgen receptor (AR) competitive binding assay that uses no animals. We validated the assay by comparing the protocols and results from other similar assays, such as the binding assay using prostate cytosol. We tested 202 natural, synthetic, and environmental chemicals that encompass a broad range of structural classes, including steroids, diethylstilbestrol and related chemicals, antiestrogens, flutamide derivatives, bisphenol A derivatives, alkylphenols, parabens, alkyloxyphenols, phthalates, siloxanes, phytoestrogens, DDTs, PCBs, pesticides, organophosphate insecticides, and other chemicals. Some of these chemicals are environmentally persistent and/or commercially important, but their AR binding affinities have not been previously reported. To the best of our knowledge, these results represent the largest and most diverse data set publicly available for chemical binding to the AR. Through a careful structure-activity relationship (SAR) examination of the data set in conjunction with knowledge of the recently reported ligand-AR crystal structures, we are able to define the general structural requirements for chemical binding to AR. Hydrophobic interactions are important for AR binding. The interaction between ligand and AR at the 3- and 17-positions of testosterone and R1881 found in other chemical classes are discussed in depth. The SAR studies of ligand binding characteristics for AR are compared to our previously reported results for estrogen receptor binding.

Androgen Receptor Antagonists↗

PubMed2003

ArrayTrack--supporting toxicogenomic research at the U.S. Food and Drug Administration National Center for Toxicological Research.

The mapping of the human genome and the determination of corresponding gene functions, pathways, and biological mechanisms are driving the emergence of the new research fields of toxicogenomics and systems toxicology. Many technological advances such as microarrays are enabling this paradigm shift that indicates an unprecedented advancement in the methods of understanding the expression of toxicity at the molecular level. At the National Center for Toxicological Research (NCTR) of the U.S. Food and Drug Administration, core facilities for genomic, proteomic, and metabonomic technologies have been established that use standardized experimental procedures to support centerwide toxicogenomic research. Collectively, these facilities are continuously producing an unprecedented volume of data. NCTR plans to develop a toxicoinformatics integrated system (TIS) for the purpose of fully integrating genomic, proteomic, and metabonomic data with the data in public repositories as well as conventional (Italic)in vitro(/Italic) and (Italic)in vivo(/Italic) toxicology data. The TIS will enable data curation in accordance with standard ontology and provide or interface a rich collection of tools for data analysis and knowledge mining. In this article the design, practical issues, and functions of the TIS are discussed through presenting its prototype version, ArrayTrack, for the management and analysis of DNA microarray data. ArrayTrack is logically constructed of three linked components: a) a library (LIB) that mirrors critical data in public databases; b) a database (MicroarrayDB) that stores microarray experiment information that is Minimal Information About a Microarray Experiment (MIAME) compliant; and c) tools (TOOL) that operate on experimental and public data for knowledge discovery. Using ArrayTrack, we can select an analysis method from the TOOL and apply the method to selected microarray data stored in the MicroarrayDB; the analysis results can be linked directly to gene information in the LIB.

Databases, Factual↗

PubMed2003

[Influence of the inhibitor of c-Met on the growth and motility of hepatocellular carcinoma cells].

OBJECTIVES: To explore the influence of c-Met inhibitor by synthetic c-Met antisense oligonucleotide, constructive c-Met antisense plasmid and the complex plasmid of U1SnRNA/ ribozyme/anti-Met on the growth and metastasis of hepatocellular carcinoma cells. METHODS: Gene transfection was operated by Lipofectin on SF7721 cells. The difference of the cells before and after transfection was compared by MTT, growth curves and transwell test in vitro. In vivo, the cells before and after transfection were implanted subcutaneously into nude mice respectively to observe tumor growth and metastasis. RESULTS: C-Met antisense oligonucleotide could inhibit the growth of hepatocellular carcinoma SF7721 cells (t=3.58, P<0.05). After transfection, the expression of c-Met protein decreased. Growth curves showed that the cells after transfection proliferated more slowly, about 50% of control cells (F=4.87, P<0.05), and their motility and invasiveness decreased, compared with those before transfected. In vivo experiment, tumors originated from c-Met antisense oligonucleotide treated cells and the antisense/ribozyme/U1SnRNA treated cells grew more slowly (about 54.5% of those from the control cells), and the latent prolonged. After 35 days, the average weight of tumors in the two group nude mice were lighter than that in the control group nude mice (F=5.17, P<0.05). CONCLUSION: Inhibition of c-Met expression by c-Met antisense oligonucleotide and the complex of antisense/ribozyme/U1SnRNA can inhibit the growth and metastasis of SF7721 hepatocarcinoma cells in vitro and in vivo.

Animals↗

PubMed2002

Prediction of estrogen receptor binding for 58,000 chemicals using an integrated system of a tree-based model with structural alerts.

A number of environmental chemicals, by mimicking natural hormones, can disrupt endocrine function in experimental animals, wildlife, and humans. These chemicals, called "endocrine-disrupting chemicals" (EDCs), are such a scientific and public concern that screening and testing 58,000 chemicals for EDC activities is now statutorily mandated. Computational chemistry tools are important to biologists because they identify chemicals most important for in vitro and in vivo studies. Here we used a computational approach with integration of two rejection filters, a tree-based model, and three structural alerts to predict and prioritize estrogen receptor (ER) ligands. The models were developed using data for 232 structurally diverse chemicals (training set) with a 10(6) range of relative binding affinities (RBAs); we then validated the models by predicting ER RBAs for 463 chemicals that had ER activity data (testing set). The integrated model gave a lower false negative rate than any single component for both training and testing sets. When the integrated model was applied to approximately 58,000 potential EDCs, 80% (approximately 46,000 chemicals) were predicted to have negligible potential (log RBA < -4.5, with log RBA = 2.0 for estradiol) to bind ER. The ability to process large numbers of chemicals to predict inactivity for ER binding and to categorically prioritize the remainder provides one biologic measure to prioritize chemicals for entry into more expensive assays (most chemicals have no biologic data of any kind). The general approach for predicting ER binding reported here may be applied to other receptors and/or reversible binding mechanisms involved in endocrine disruption.

Animals↗

PubMedDate not supplied

Decision forest: combining the predictions of multiple independent decision tree models.

The techniques of combining the results of multiple classification models to produce a single prediction have been investigated for many years. In earlier applications, the multiple models to be combined were developed by altering the training set. The use of these so-called resampling techniques, however, poses the risk of reducing predictivity of the individual models to be combined and/or over fitting the noise in the data, which might result in poorer prediction of the composite model than the individual models. In this paper, we suggest a novel approach, named Decision Forest, that combines multiple Decision Tree models. Each Decision Tree model is developed using a unique set of descriptors. When models of similar predictive quality are combined using the Decision Forest method, quality compared to the individual models is consistently and significantly improved in both training and testing steps. An example will be presented for prediction of binding affinity of 232 chemicals to the estrogen receptor.

Algorithms↗

PubMedDate not supplied

Models of steroid binding based on the minimum deviation of structurally assigned 13C NMR spectra analysis (MiDSASA).

This paper develops a quantitative k-nearest neighbors modeling technique. The technique is used to demonstrate that a compound's biological binding activity to a receptor can be calculated from the minimum of the square root of the sum of squared deviations (SSSD) of a structurally assigned chemical shift on a template between the unknown compound to be predicted and a set of known compounds with known activities. When building models of biological activity, nonlinear relationships are built into the input training data. If a model is developed by selecting only compounds with minimum structurally assigned chemical shift deviations from the unknown compound, some of the nonlinear relationships can be removed. The smaller the total chemical shift deviation between a compound with known activity and another compound with unknown activity, the more likely it will have similar biological, chemical, and physical properties. This means that a model can be produced without rigorous statistics or neural networks. This technique is similar to structure-activity relationship (SAR) modeling, but instead of relying on substructure fragments to produce a model, this new model is based on minimum chemical shift differences on those substructure fragments. We refer to this method as minimum deviation of structurally assigned spectra analysis (MiDSASA) modeling. Modeling by the minimum deviation concept can be applied to other chemoinformatic data analyses such as metabolite concentrations in metabolic pathways for metabolomics research. A MiDSASA template model for 30 steroids binding the corticosterone binding globulin based on the activity factors of the two nearest compounds had a correlation of 0.88. A MiDSASA template model for 50 steroids binding the aromatse enzyme based on the average activity of the four nearest compounds had a correlation of 0.71.

Aromatase↗