Performance of random forests and logic regression methods using mini-exome sequence data.
Ontology highlight
ABSTRACT:
SUBMITTER: Kim Y
PROVIDER: S-EPMC3287827 | biostudies-other | 2011 Nov
REPOSITORIES: biostudies-other
Publications
Performance of random forests and logic regression methods using mini-exome sequence data.
Kim Yoonhee Y Li Qing Q Cropp Cheryl D CD Sung Heejong H Cai Juanliang J Simpson Claire L CL Perry Brian B Dasgupta Abhijit A Malley James D JD Wilson Alexander F AF Bailey-Wilson Joan E JE
BMC proceedings 20111129
Machine learning approaches are an attractive option for analyzing large-scale data to detect genetic variants that contribute to variation of a quantitative trait, without requiring specific distributional assumptions. We evaluate two machine learning methods, random forests and logic regression, and compare them to standard simple univariate linear regression, using the Genetic Analysis Workshop 17 mini-exome data. We also apply these methods after collapsing multiple rare variants within gene ...[more]
Similar Datasets
Project description:BACKGROUND: In bioprocess development, the needs of data analysis include (1) getting overview to existing data sets, (2) identifying primary control parameters, (3) determining a useful control direction, and (4) planning future experiments. In particular, the integration of multiple data sets causes that these needs cannot be properly addressed by regression models that assume linear input-output relationship or unimodality of the response function. Regularized regression and random forests, on the other hand, have several properties that may appear important in this context. They are capable, e.g., in handling small number of samples with respect to the number of variables, feature selection, and the visualization of response surfaces in order to present the prediction results in an illustrative way. RESULTS: In this work, the applicability of regularized regression (Lasso) and random forests (RF) in bioprocess data mining was examined, and their performance was benchmarked against multiple linear regression. As an example, we used data from a culture media optimization study for microbial hydrogen production. All the three methods were capable in providing a significant model when the five variables of the culture media optimization were linearly included in modeling. However, multiple linear regression failed when also the multiplications and squares of the variables were included in modeling. In this case, the modeling was still successful with Lasso (correlation between the observed and predicted yield was 0.69) and RF (0.91). CONCLUSION: We found that both regularized regression and random forests were able to produce feasible models, and the latter was efficient in capturing the non-linearity in the data. In this kind of a data mining task of bioprocess data, both methods outperform multiple linear regression.
| S-EPMC3750505 | biostudies-literature
Project description:Environmental data may be "large" due to number of records, number of covariates, or both. Random forests has a reputation for good predictive performance when using many covariates with nonlinear relationships, whereas spatial regression, when using reduced rank methods, has a reputation for good predictive performance when using many records that are spatially autocorrelated. In this study, we compare these two techniques using a data set containing the macroinvertebrate multimetric index (MMI) at 1859 stream sites with over 200 landscape covariates. A primary application is mapping MMI predictions and prediction errors at 1.1 million perennial stream reaches across the conterminous United States. For the spatial regression model, we develop a novel transformation procedure that estimates Box-Cox transformations to linearize covariate relationships and handles possibly zero-inflated covariates. We find that the spatial regression model with transformations, and a subsequent selection of significant covariates, has cross-validation performance comparable to random forests. We also find that prediction interval coverage is close to nominal for each method, but that spatial regression prediction intervals tend to be narrower and have less variability than quantile regression forest prediction intervals. A simulation study is used to generalize results and clarify advantages of each modeling approach.
| S-EPMC7089425 | biostudies-literature
Project description:Machine learning methods and in particular random forests are promising approaches for prediction based on high dimensional omics data sets. They provide variable importance measures to rank predictors according to their predictive power. If building a prediction model is the main goal of a study, often a minimal set of variables with good prediction performance is selected. However, if the objective is the identification of involved variables to find active networks and pathways, approaches that aim to select all relevant variables should be preferred. We evaluated several variable selection procedures based on simulated data as well as publicly available experimental methylation and gene expression data. Our comparison included the Boruta algorithm, the Vita method, recurrent relative variable importance, a permutation approach and its parametric variant (Altmann) as well as recursive feature elimination (RFE). In our simulation studies, Boruta was the most powerful approach, followed closely by the Vita method. Both approaches demonstrated similar stability in variable selection, while Vita was the most robust approach under a pure null model without any predictor variables related to the outcome. In the analysis of the different experimental data sets, Vita demonstrated slightly better stability in variable selection and was less computationally intensive than Boruta. In conclusion, we recommend the Boruta and Vita approaches for the analysis of high-dimensional data sets. Vita is considerably faster than Boruta and thus more suitable for large data sets, but only Boruta can also be applied in low-dimensional settings.
| S-EPMC6433899 | biostudies-literature
Project description:Understanding and identifying the markers and clinical information that are associated with colorectal cancer (CRC) patient survival is needed for early detection and diagnosis. In this work, we aimed to build a simple model using Cox proportional hazards (PH) and random survival forest (RSF) and find a robust signature for predicting CRC overall survival. We used stepwise regression to develop Cox PH model to analyse 54 common differentially expressed genes from three mutations. RSF is applied using log-rank and log-rank-score based on 5000 survival trees, and therefore, variables important obtained to find the genes that are most influential for CRC survival. We compared the predictive performance of the Cox PH model and RSF for early CRC detection and diagnosis. The results indicate that SLC9A8, IER5, ARSJ, ANKRD27, and PIPOX genes were significantly associated with the CRC overall survival. In addition, age, sex, and stages are also affecting the CRC overall survival. The RSF model using log-rank is better than log-rank-score, while log-rank-score needed more trees to stabilize. Overall, the imputation of missing values enhanced the model's predictive performance. In addition, Cox PH predictive performance was better than RSF.
| S-EPMC8716055 | biostudies-literature
Project description:Identifying genetic variants associated with complex disease in high-dimensional data is a challenging problem, and complicated etiologies such as gene-gene interactions are often ignored in analyses. The data-mining method Random Forests (RF) can handle high-dimensions; however, in high-dimensional data, RF is not an effective filter for identifying risk factors associated with the disease trait via complex genetic models such as gene-gene interactions without strong marginal components. Here we propose an extension called Weighted Random Forests (wRF), which incorporates tree-level weights to emphasize more accurate trees in prediction and calculation of variable importance. We demonstrate through simulation and application to data from a genetic study of addiction that wRF can outperform RF in high-dimensional data, although the improvements are modest and limited to situations with effect sizes that are larger than what is realistic in genetics of complex disease. Thus, the current implementation of wRF is unlikely to improve detection of relevant predictors in high-dimensional genetic data, but may be applicable in other situations where larger effect sizes are anticipated.
| S-EPMC3912194 | biostudies-literature
Project description:BACKGROUND:In the last years more and more multi-omics data are becoming available, that is, data featuring measurements of several types of omics data for each patient. Using multi-omics data as covariate data in outcome prediction is both promising and challenging due to the complex structure of such data. Random forest is a prediction method known for its ability to render complex dependency patterns between the outcome and the covariates. Against this background we developed five candidate random forest variants tailored to multi-omics covariate data. These variants modify the split point selection of random forest to incorporate the block structure of multi-omics data and can be applied to any outcome type for which a random forest variant exists, such as categorical, continuous and survival outcomes. Using 20 publicly available multi-omics data sets with survival outcome we compared the prediction performances of the block forest variants with alternatives. We also considered the common special case of having clinical covariates and measurements of a single omics data type available. RESULTS:We identify one variant termed "block forest" that outperformed all other approaches in the comparison study. In particular, it performed significantly better than standard random survival forest (adjusted p-value: 0.027). The two best performing variants have in common that the block choice is randomized in the split point selection procedure. In the case of having clinical covariates and a single omics data type available, the improvements of the variants over random survival forest were larger than in the case of the multi-omics data. The degrees of improvements over random survival forest varied strongly across data sets. Moreover, considering all clinical covariates mandatorily improved the performance. This result should however be interpreted with caution, because the level of predictive information contained in clinical covariates depends on the specific application. CONCLUSIONS:The new prediction method block forest for multi-omics data can significantly improve the prediction performance of random forest and outperformed alternatives in the comparison. Block forest is particularly effective for the special case of using clinical covariates in combination with measurements of a single omics data type.
| S-EPMC6598279 | biostudies-literature
Project description:BackgroundDiabetic retinopathy (DR) is one of the leading causes of blindness in the United States and world-wide. DR is a silent disease that may go unnoticed until it is too late for effective treatment. Therefore, early detection could improve the chances of therapeutic interventions that would alleviate its effects.MethodologyGraded fundus photography and systemic data from 3443 ACCORD-Eye Study participants were used to estimate Random Forest (RF) and logistic regression classifiers. We studied the impact of sample size on classifier performance and the possibility of using RF generated class conditional probabilities as metrics describing DR risk. RF measures of variable importance are used to detect factors that affect classification performance.Principal findingsBoth types of data were informative when discriminating participants with or without DR. RF based models produced much higher classification accuracy than those based on logistic regression. Combining both types of data did not increase accuracy but did increase statistical discrimination of healthy participants who subsequently did or did not have DR events during four years of follow-up. RF variable importance criteria revealed that microaneurysms counts in both eyes seemed to play the most important role in discrimination among the graded fundus variables, while the number of medicines and diabetes duration were the most relevant among the systemic variables.Conclusions and significanceWe have introduced RF methods to DR classification analyses based on fundus photography data. In addition, we propose an approach to DR risk assessment based on metrics derived from graded fundus photography and systemic data. Our results suggest that RF methods could be a valuable tool to diagnose DR diagnosis and evaluate its progression.
| S-EPMC4062420 | biostudies-literature
Project description:Random forests (RF) is a powerful species distribution model (SDM) algorithm. This ensemble model by default can produce categorical and numerical species distribution maps based on its classification tree (CT) and regression tree (RT) algorithms, respectively. The CT algorithm can also produce numerical predictions (class probability). Here, we present a detailed procedure involving the use of the CT and RT algorithms using the RF method with presence-only data to model the distribution of species. CT and RT are used to generate numerical prediction maps, and then numerical predictions are converted to binary predictions through objective threshold-setting methods. We also applied simple methods to deal with collinearity of predictor variables and spatial autocorrelation of species occurrence data. A geographically stratified sampling method was employed for generating pseudo-absences. The detailed procedural framework is meant to be a generic method to be applied to virtually any SDM prediction question using presence-only data. •How to use RF as a standard method for generic species distributions with presence-only data•How to choose RF (CT or RT) methods for the distribution modeling of species•A general and detailed procedure for any SDM prediction question.
| S-EPMC6812352 | biostudies-literature
Project description:Extending previous work on quantile classifiers (q-classifiers) we propose the q*-classifier for the class imbalance problem. The classifier assigns a sample to the minority class if the minority class conditional probability exceeds 0 < q* < 1, where q* equals the unconditional probability of observing a minority class sample. The motivation for q*-classification stems from a density-based approach and leads to the useful property that the q*-classifier maximizes the sum of the true positive and true negative rates. Moreover, because the procedure can be equivalently expressed as a cost-weighted Bayes classifier, it also minimizes weighted risk. Because of this dual optimization, the q*-classifier can achieve near zero risk in imbalance problems, while simultaneously optimizing true positive and true negative rates. We use random forests to apply q*-classification. This new method which we call RFQ is shown to outperform or is competitive with existing techniques with respect to tt-mean performance and variable selection. Extensions to the multiclass imbalanced setting are also considered.
| S-EPMC6370055 | biostudies-literature
Project description:BACKGROUND: High-throughput sequencing is rapidly becoming common practice in clinical diagnosis and cancer research. Many algorithms have been developed for somatic single nucleotide variant (SNV) detection in matched tumor-normal DNA sequencing. Although numerous studies have compared the performance of various algorithms on exome data, there has not yet been a systematic evaluation using PCR-enriched amplicon data with a range of variant allele fractions. The recently developed gold standard variant set for the reference individual NA12878 by the NIST-led "Genome in a Bottle" Consortium (NIST-GIAB) provides a good resource to evaluate admixtures with various SNV fractions. RESULTS: Using the NIST-GIAB gold standard, we compared the performance of five popular somatic SNV calling algorithms (GATK UnifiedGenotyper followed by simple subtraction, MuTect, Strelka, SomaticSniper and VarScan2) for matched tumor-normal amplicon and exome sequencing data. CONCLUSIONS: We demonstrated that the five commonly used somatic SNV calling methods are applicable to both targeted amplicon and exome sequencing data. However, the sensitivities of these methods vary based on the allelic fraction of the mutation in the tumor sample. Our analysis can assist researchers in choosing a somatic SNV calling method suitable for their specific needs.
| S-EPMC3986649 | biostudies-literature