A novel algorithm for simplification of complex gene classifiers in cancer.
Ontology highlight
ABSTRACT:
SUBMITTER: Wilson RA
PROVIDER: S-EPMC4233649 | biostudies-literature | 2013 Sep
REPOSITORIES: biostudies-literature
Publications
A novel algorithm for simplification of complex gene classifiers in cancer.
Wilson Raphael A RA Teng Ling L Bachmeyer Karen M KM Bissonnette Mei Lin Z ML Husain Aliya N AN Parham David M DM Triche Timothy J TJ Wing Michele R MR Gastier-Foster Julie M JM Barr Frederic G FG Hawkins Douglas S DS Anderson James R JR Skapek Stephen X SX Volchenboum Samuel L SL
Cancer research 20130802 18
The clinical application of complex molecular classifiers as diagnostic or prognostic tools has been limited by the time and cost needed to apply them to patients. Using an existing 50-gene expression signature known to separate two molecular subtypes of the pediatric cancer rhabdomyosarcoma, we show that an exhaustive iterative search algorithm can distill this complex classifier down to two or three features with equal discrimination. We validated the two-gene signatures using three separate a ...[more]
Similar Datasets
Project description:MOTIVATION:Expert-labeled data are essential to train phenotyping algorithms for cohort identification. However expert labeling is time and labor intensive, and the costs remain prohibitive for scaling phenotyping to wider use-cases. RESULTS:We present an approach referred to as polar labeling (PL), to create silver standard for training machine learning (ML) for disease classification. We test the hypothesis that ML models trained on the silver standard created by applying PL on unlabeled patient records, are comparable in performance to the ML models trained on gold standard, created by clinical experts through manual review of patient records. We perform experimental validation using health records of 38 023 patients spanning six diseases. Our results demonstrate the superior performance of the proposed approach. AVAILABILITY AND IMPLEMENTATION:We provide a Python implementation of the algorithm and the Python code developed for this study on Github. SUPPLEMENTARY INFORMATION:Supplementary data are available at Bioinformatics online.
| S-EPMC7214041 | biostudies-literature
Project description:In order to elucidate the biological variance between normal ovarian surface epithelial (NOSE) and epithelial ovarian cancer (EOC) cells, and to build a molecular classifier to discover new markers distinguishing these cells, we analysed gene expression patterns of 65 primary cultures of these tissues by oligonucleotide microarray. Unsupervised clustering highlights three subgroups of tumours: low malignant potential tumours, invasive solid tumours and tumour cells derived from ascites. We selected 18 genes with expression profiles that enable the distinction of NOSE from these three groups of EOC with 92% accuracy. Validation using an independent published data set derived from tissues or primary cultures confirmed a high accuracy (87-96%). The distinctive expression pattern of a subset of genes was validated by quantitative reverse transcription-PCR. An ovarian-specific tissue array representing tissues from NOSE and EOC samples of various subtypes and grades was used to further assess the protein expression patterns of two differentially expressed genes (Msln and BMP-2) by immunohistochemistry. This study highlights the relevance of using primary cultures of epithelial ovarian cells as a model system for gene profiling studies and demonstrates that the statistical analysis of gene expression profiling is a useful approach for selecting novel molecular tumour markers.
| S-EPMC2361148 | biostudies-literature
Project description:BackgroundImprovement of patient quality of life is the ultimate goal of biomedical research, particularly when dealing with complex, chronic and debilitating conditions such as inflammatory bowel disease (IBD). This is largely dependent on receiving an accurate and rapid diagnose, an effective treatment and in the prediction and prevention of side effects and complications. The low sensitivity and specificity of current markers burden their general use in the clinical practice. New biomarkers with accurate predictive ability are needed to achieve a personalized approach that take the inter-individual differences into consideration.MethodsWe performed a high throughput approach using microarray gene expression profiling of colon pinch biopsies from IBD patients to identify predictive transcriptional signatures associated with intestinal inflammation, differential diagnosis (Crohn's disease or ulcerative colitis), response to glucocorticoids (resistance and dependence) or prognosis (need for surgery). Class prediction was performed with self-validating Prophet software package.ResultsTranscriptional profiling divided patients in two subgroups that associated with degree of inflammation. Class predictors were identified with predictive accuracy ranging from 67 to 100%. The expression accuracy was confirmed by real time-PCR quantification. Functional analysis of the predictor genes showed that they play a role in immune responses to bacteria (PTN, OLFM4 and LILRA2), autophagy and endocytocis processes (ATG16L1, DNAJC6, VPS26B, RABGEF1, ITSN1 and TMEM127) and glucocorticoid receptor degradation (STS and MMD2).ConclusionsWe conclude that using analytical algorithms for class prediction discovery can be useful to uncover gene expression profiles and identify classifier genes with potential stratification utility of IBD patients, a major step towards personalized therapy.
| S-EPMC3796518 | biostudies-literature
Project description:Automated protein annotation using the Gene Ontology (GO) plays an important role in the biosciences. Evaluation has always been considered central to developing novel annotation methods, but little attention has been paid to the evaluation metrics themselves. Evaluation metrics define how well an annotation method performs and allows for them to be ranked against one another. Unfortunately, most of these metrics were adopted from the machine learning literature without establishing whether they were appropriate for GO annotations. We propose a novel approach for comparing GO evaluation metrics called Artificial Dilution Series (ADS). Our approach uses existing annotation data to generate a series of annotation sets with different levels of correctness (referred to as their signal level). We calculate the evaluation metric being tested for each annotation set in the series, allowing us to identify whether it can separate different signal levels. Finally, we contrast these results with several false positive annotation sets, which are designed to expose systematic weaknesses in GO assessment. We compared 37 evaluation metrics for GO annotation using ADS and identified drastic differences between metrics. We show that some metrics struggle to differentiate between different signal levels, while others give erroneously high scores to the false positive data sets. Based on our findings, we provide guidelines on which evaluation metrics perform well with the Gene Ontology and propose improvements to several well-known evaluation metrics. In general, we argue that evaluation metrics should be tested for their performance and we provide software for this purpose (https://bitbucket.org/plyusnin/ads/). ADS is applicable to other areas of science where the evaluation of prediction results is non-trivial.
| S-EPMC6855565 | biostudies-literature
Project description:Conventional point cloud simplification algorithms have problems including nonuniform simplification, a deficient reflection of point cloud characteristics, unreasonable weight distribution, and high computational complexity. A simplification algorithm, namely, the multi-index weighting simplification algorithm (MIWSA), is proposed in this paper. First, the point cloud is organized with a bounding box and kd-trees to find the neighborhood of each point, and the points are divided into small segments. Second, the feature index of each point is calculated to indicate the characteristics of the points. Third, the analytic hierarchy process (AHP) and criteria importance through intercriteria correlation (CRITIC) are applied to weight these indexes to determine whether each point is a feature point. Fourth, non-feature points are judged as saved or abandoned according to their spatial relationship with the feature points. To verify the effect of the MIWSA, 3D model scanning datasets are calculated and analyzed, as well as field area scanning datasets. The accuracy for the 3D model scanning datasets is assessed by the surface area and patch numbers of the encapsulated surfaces, and that for field area scanning datasets is evaluated by the DEM error statistics. Compared with existing algorithms, the overall accuracy of the MIWSA is 5% to 15% better. Additionally, the running time is shorter than most. The experimental results illustrate that the MIWSA can simplify point clouds more precisely and uniformly.
| S-EPMC9573161 | biostudies-literature
Project description:Monoclonal antibodies (mAbs) exhibit biexponential profiles in plasma that are commonly described with a standard two-compartment model with elimination from the central compartment. These models adequately describe mAb plasma PK. However, these models ignore elimination from the peripheral compartment. This may lead to underestimation of the volume of distribution of the peripheral compartment and thus over-predicts concentration in the peripheral compartment. We developed a simple and physiologically relevant model that incorporates information on binding and dissociation rates between mAb and FcRn receptor, mAb uptake, reflection, and catabolic degradation. We employed a previously published PBPK model and, with assumptions regarding rates of processes controlling mAb disposition, reduced the complex PBPK model to a simpler circular model with central, peripheral, and lymph compartments specifying elimination from both central and peripheral. We successfully applied the model to describe the PK of an investigational mAb. Our model presents an improvement over standard two-compartmental models in predicting whole-body average tissue concentrations while adequately describing plasma PK with minimal complexity and physiologically more meaningful parameters.
| S-EPMC4070266 | biostudies-literature
Project description:The abundance and functional orientation of tumor-infiltrating lymphocytes in breast cancer is associated with distant metastasis-free survival, yet how this association is influenced by tumor phenotypic heterogeneity is poorly understood. Here, a bioinformatics approach defined tumor biologic attributes that influence this association and delineated tumor subtypes that may differ in their ability to sustain durable antitumor immune responses. A large database of breast tumor expression profiles and associated clinical data was compiled, from which the ability of phenotypic markers to significantly influence the prognostic performance of a classification model that incorporates immune cell-specific gene signatures was ascertained. Markers of cell proliferation and intrinsic molecular subtype reproducibly distinguished two breast cancer subtypes that we refer to as immune benefit-enabled (IBE) and immune benefit-disabled (IBD). The IBE tumors, comprised mostly of highly proliferative tumors of the basal-like, HER2-enriched, and luminal B subtypes, could be stratified by the immune classifier into significantly different prognostic groups, while IBD tumors could not, indicating the potential for productive engagement of metastasis-protective immunity in IBE tumors, but not in IBD tumors. The prognostic stratification in IBE was independent of conventional variables. Gene network analysis predicted the activation of TNFα/IFNγ signaling pathways in IBE tumors and the activation of the transforming growth factor-β pathway in IBD tumors. This prediction supports a model in which breast tumors can be distinguished on the basis of their potential for metastasis-protective immune responsiveness. Whether IBE and IBD represent clinically relevant contexts for evaluating sensitivity to immunotherapeutic agents warrants further investigation. Cancer Immunol Res; 4(7); 600-10. ©2016 AACR.
| S-EPMC4930674 | biostudies-literature
Project description:BACKGROUND: Protein complexes are basic cellular entities that carry out the functions of their components. It can be found that in databases of protein complexes of yeast like CYC2008, the major type of known protein complexes is heterodimeric complexes. Although a number of methods for trying to predict sets of proteins that form arbitrary types of protein complexes simultaneously have been proposed, it can be found that they often fail to predict heterodimeric complexes. RESULTS: In this paper, we have designed several features characterizing heterodimeric protein complexes based on genomic data sets, and proposed a supervised-learning method for the prediction of heterodimeric protein complexes. This method learns the parameters of the features, which are embedded in the naïve Bayes classifier. The log-likelihood ratio derived from the naïve Bayes classifier with the parameter values obtained by maximum likelihood estimation gives the score of a given pair of proteins to predict whether the pair is a heterodimeric complex or not. A five-fold cross-validation shows good performance on yeast. The trained classifiers also show higher predictability than various existing algorithms on yeast data sets with approximate and exact matching criteria. CONCLUSIONS: Heterodimeric protein complex prediction is a rather harder problem than heteromeric protein complex prediction because heterodimeric protein complex is topologically simpler. However, it turns out that by designing features specialized for heterodimeric protein complexes, predictability of them can be improved. Thus, the design of more sophisticate features for heterodimeric protein complexes as well as the accumulation of more accurate and useful genome-wide data sets will lead to higher predictability of heterodimeric protein complexes. Our tool can be downloaded from http://imi.kyushu-u.ac.jp/~om/.
| S-EPMC4219333 | biostudies-other
Project description:BackgroundClustering DNA sequences into functional groups is an important problem in bioinformatics. We propose a new alignment-free algorithm, mBKM, based on a new distance measure, DMk, for clustering gene sequences. This method transforms DNA sequences into the feature vectors which contain the occurrence, location and order relation of k-tuples in DNA sequence. Afterwards, a hierarchical procedure is applied to clustering DNA sequences based on the feature vectors.ResultsThe proposed distance measure and clustering method are evaluated by clustering functionally related genes and by phylogenetic analysis. This method is also compared with BlastClust, CD-HIT-EST and some others. The experimental results show our method is effective in classifying DNA sequences with similar biological characteristics and in discovering the underlying relationship among the sequences.ConclusionsWe introduced a novel clustering algorithm which is based on a new sequence similarity measure. It is effective in classifying DNA sequences with similar biological characteristics and in discovering the relationship among the sequences.
| S-EPMC3443659 | biostudies-literature
Project description:Cancer and healthy cells have distinct distributions of molecular properties and thus respond differently to drugs. Cancer drugs ideally kill cancer cells while limiting harm to healthy cells. However, the inherent variance among cells in both cancer and healthy cell populations increases the difficulty of selective drug action. Here we formalize a classification framework based on the idea that an ideal cancer drug should maximally discriminate between cancer and healthy cells. More specifically, this discrimination should be performed on the basis of measurable cell markers. We divide the problem into three parts which we explore with examples. First, molecular markers should discriminate cancer cells from healthy cells at the single-cell level. Second, the effects of drugs should be statistically predicted by these molecular markers. Third, drugs should be optimized for classification performance. We find that expression levels of a handful of genes suffice to discriminate well between individual cells in cancer and healthy tissue. We also find that gene expression predicts the efficacy of some cancer drugs, suggesting that these cancer drugs act as suboptimal classifiers using gene profiles. Finally, we formulate a framework that defines an optimal drug, and predicts drug cocktails that may target cancer more accurately than the individual drugs alone. Conceptualizing cancer drugs as solving a discrimination problem in the high-dimensional space of molecular markers promises to inform the design of new cancer drugs and drug cocktails.
| S-EPMC4172566 | biostudies-literature