Systematic Selection of Age-Associated mRNA Markers and the Development of Predicted Models for Forensic Age Inference by Three Machine Learning Methods
Ontology highlight
ABSTRACT:
SUBMITTER: Jin X
PROVIDER: S-EPMC9283997 | biostudies-literature |
REPOSITORIES: biostudies-literature
Similar Datasets
Project description:Many modern problems in medicine and public health leverage machine-learning methods to predict outcomes based on observable covariates. In a wide array of settings, predicted outcomes are used in subsequent statistical analysis, often without accounting for the distinction between observed and predicted outcomes. We call inference with predicted outcomes postprediction inference. In this paper, we develop methods for correcting statistical inference using outcomes predicted with arbitrarily complicated machine-learning models including random forests and deep neural nets. Rather than trying to derive the correction from first principles for each machine-learning algorithm, we observe that there is typically a low-dimensional and easily modeled representation of the relationship between the observed and predicted outcomes. We build an approach for postprediction inference that naturally fits into the standard machine-learning framework where the data are divided into training, testing, and validation sets. We train the prediction model in the training set, estimate the relationship between the observed and predicted outcomes in the testing set, and use that relationship to correct subsequent inference in the validation set. We show our postprediction inference (postpi) approach can correct bias and improve variance estimation and subsequent statistical inference with predicted outcomes. To show the broad range of applicability of our approach, we show postpi can improve inference in two distinct fields: modeling predicted phenotypes in repurposed gene expression data and modeling predicted causes of death in verbal autopsy data. Our method is available through an open-source R package: https://github.com/leekgroup/postpi.
| S-EPMC7720220 | biostudies-literature
Project description:Current methods for pregnancy diagnostics are based on immunodetection of pregnancy-specific proteins and in a forensic context suffer from sensitivity and specificity issues. Here, we applied reverse transcriptase polymerase chain reaction (RT-PCR) technology to 11 genes previously reported with placental mRNA circulating in maternal blood. We found two genes, hPL and betahCG, with pregnancy-specific expression in whole blood samples. RT-PCR detection of hPL was positive in all samples tested throughout the pregnancy, whereas betahCG was detectable until half of the second trimester but not at later gestation ages. For hPL, in vitro stability of the transcript was demonstrated until 2 months of age, and the hPL-specific RT-PCR assay applied was highly sensitive with reliable detection from down to 0.25 cm(2) dried bloodstain. We therefore suggest hPL-specific RT-PCR as a new molecular tool for forensic pregnancy diagnostics from dried blood stains. Moreover, our results indicate that the time-wise reverse expression of hPL and betahCG during pregnancy may allow an RT-PCR-based estimation of the gestational age from blood stains, adding to the value of forensic pregnancy diagnosis for crime scene investigations.
| S-EPMC2795858 | biostudies-literature
Project description:Genomic selection (GS) has been used to optimize genetic gains when phenotypic selection is considered costly and difficult to measure. The objective of this work was to evaluate the efficiency and consistency of GS prediction for cassava yield traits (Manihot esculenta Crantz) using different methods, taking into account the effect of population structure. BLUPs and deregressed BLUPs were obtained for 888 cassava accessions and evaluated for fresh root yield, dry root yield and dry matter content in roots in 21 trials conducted from 2011 to 2016. The deregressed BLUPs obtained for the accessions from a 48K single nucleotide polymorphism dataset were used for genomic predictions based on the BayesB, BLASSO, RR-BLUP, G-BLUP and RKHS methods. The accessions' BLUPs were used in the validation step using four cross-validation strategies, taking into account population structure and different GS methods. Similar estimates of predictive ability and bias were identified for the different genomic selection methods in the first cross-validation strategy. Lower predictive ability was observed for fresh root yield (0.4569 -RR-BLUP to 0.4756-RKHS) and dry root yield (0.4689 -G-BLUP to 0.4818-RKHS) in comparison with dry matter content (0.5655 -BLASSO to 0.5670 -RKHS). However, the RKHS method exhibited higher efficiency and consistency in most of the validation scenarios in terms of prediction ability for fresh root yield and dry root yield. The correlations of the genomic estimated breeding values between the genomic selection methods were quite high (0.99-1.00), resulting in high coincidence of clone selection regardless of the genomic selection method. The deviance analyses within and between the validation clusters formed by the discriminant analysis of principal components were significant for all traits. Therefore, this study indicated that i) the prediction of dry matter content was more accurate compared to that of yield traits, possibly as a result of the smaller influence of non-additive genetic effects; ii) the RKHS method resulted in high and stable prediction ability in most of the validation scenarios; and iii) some kinship between the validation and training populations is desirable in order for genomic selection to succeed due to the significant effect of population structure on genomic selection predictions.
| S-EPMC6855463 | biostudies-literature
Project description:Aging is a complicated process characterized by progressive and extensive changes in physiological homeostasis at the organismal, tissue, and cellular levels. In modern society, age estimation is essential in a large variety of legal rights and duties. Accumulating evidence suggests roles for microRNAs (miRNAs) and circular RNAs (circRNAs) in regulating numerous processes during aging. Here, we performed circRNA sequencing in two age groups and analyzed microarray data of 171 healthy subjects (17-104 years old) downloaded from Gene Expression Omnibus (GEO) and ArrayExpress databases with integrated bioinformatics methods. A total of 1,403 circular RNAs were differentially expressed between young and old groups, and 141 circular RNAs were expressed exclusively in elderly samples while 10 circular RNAs were expressed only in young subjects. Based on their expression pattern in these two groups, the circular RNAs were categorized into three classes: age-related expression between young and old, age-limited expression-young only, and age-limited expression-old only. Top five expressed circular RNAs among three classes and a total of 18 differentially expressed microRNAs screened from online databases were selected to validate using RT-qPCR tests. An independent set of 200 blood samples (20-80 years old) was used to develop age prediction models based on 15 age-related noncoding RNAs (11 microRNAs and 4 circular RNAs). Different machine learning algorithms for age prediction were applied, including regression tree, bagging, support vector regression (SVR), random forest regression (RFR), and XGBoost. Among them, random forest regression model performed best in both training set (mean absolute error = 3.68 years, r = 0.96) and testing set (MAE = 6.840 years, r = 0.77). Models using one single type of predictors, circular RNAs-only or microRNAs-only, result in bigger errors. Smaller prediction errors were shown in males than females when constructing models according to different-sex separately. Putative microRNA targets (430 genes) were enriched in the cellular senescence pathway and cell homeostasis and cell differentiation regulation, indirectly indicating that the microRNAs screened in our study were correlated with development and aging. This study demonstrates that the noncoding RNA aging clock has potential in predicting chronological age and will be an available biological marker in routine forensic investigation to predict the age of biological samples.
| S-EPMC9732945 | biostudies-literature
Project description:BackgroundMolecular biology is currently facing the challenging task of functionally characterizing the proteome. The large number of possible protein-protein interactions and complexes, the variety of environmental conditions and cellular states in which these interactions can be reorganized, and the multiple ways in which a protein can influence the function of others, requires the development of experimental and computational approaches to analyze and predict functional associations between proteins as part of their activity in the interactome.Methodology/principal findingsWe have studied the possibility of constructing a classifier in order to combine the output of the several protein interaction prediction methods. The AODE (Averaged One-Dependence Estimators) machine learning algorithm is a suitable choice in this case and it provides better results than the individual prediction methods, and it has better performances than other tested alternative methods in this experimental set up. To illustrate the potential use of this new AODE-based Predictor of Protein InterActions (APPIA), when analyzing high-throughput experimental data, we show how it helps to filter the results of published High-Throughput proteomic studies, ranking in a significant way functionally related pairs.AvailabilityAll the predictions of the individual methods and of the combined APPIA predictor, together with the used datasets of functional associations are available at http://ecid.bioinfo.cnio.es/.ConclusionsWe propose a strategy that integrates the main current computational techniques used to predict functional associations into a unified classifier system, specifically focusing on the evaluation of poorly characterized protein pairs. We selected the AODE classifier as the appropriate tool to perform this task. AODE is particularly useful to extract valuable information from large unbalanced and heterogeneous data sets. The combination of the information provided by five prediction interaction prediction methods with some simple sequence features in APPIA is useful in establishing reliability values and helpful to prioritize functional interactions that can be further experimentally characterized.
| S-EPMC2848617 | biostudies-literature
Project description:BackgroundLocal structures of target mRNAs play a significant role in determining the efficacies of antisense oligonucleotides (ODNs), but some structure-based target site selection methods are limited by uncertainties in RNA secondary structure prediction. If all the predicted structures of a given mRNA within a certain energy limit could be used simultaneously, target site selection would obviously be improved in both reliability and efficiency. In this study, some key problems in ODN target selection on the basis of multiple predicted target mRNA structures are systematically discussed.ResultsTwo methods were considered for merging topologically different RNA structures into integrated representations. Several parameters were derived to characterize local target site structures. Statistical analysis on a dataset with 448 ODNs against 28 different mRNAs revealed 9 features quantitatively associated with efficacy. Features of structural consistency seemed to be more highly correlated with efficacy than indices of the proportion of bases in single-stranded or double-stranded regions. The local structures of the target site 5' and 3' termini were also shown to be important in target selection. Neural network efficacy predictors using these features, defined on integrated structures as inputs, performed well in "minus-one-gene" cross-validation experiments.ConclusionTopologically different target mRNA structures can be merged into integrated representations and then used in computer-aided ODN design. The results of this paper imply that some features characterizing multiple predicted target site structures can be used to predict ODN efficacy.
| S-EPMC1421440 | biostudies-literature
Project description:Objectives: In this study, we applied the random forest (RF) algorithm to birth-cohort data to train a model to predict low cognitive ability at 5 years of age and to identify the important predictive features. Methods: Data was from 1,070 participants in the Irish population-based BASELINE cohort. A RF model was trained to predict an intelligence quotient (IQ) score ≤90 at age 5 years using maternal, infant, and sociodemographic features. Feature importance was examined and internal validation performed using 10-fold cross validation repeated 5 times. Results The five most important predictive features were the total years of maternal schooling, infant Apgar score at 1 min, socioeconomic index, maternal BMI, and alcohol consumption in the first trimester. On internal validation a parsimonious RF model based on 11 features showed excellent predictive ability, correctly classifying 95% of participants. This provides a foundation suitable for external validation in an unseen cohort. Conclusion: Machine learning approaches to large existing datasets can provide accurate feature selection to improve risk prediction. Further validation of this model is required in cohorts representative of the general population.
| S-EPMC9684182 | biostudies-literature
Project description:Different types of protein-protein interactions make different protein-protein interface patterns. Different machine learning methods are suitable to deal with different types of data. Then, is it the same situation that different interface patterns are preferred for prediction by different machine learning methods? Here, four different machine learning methods were employed to predict protein-protein interface residue pairs on different interface patterns. The performances of the methods for different types of proteins are different, which suggest that different machine learning methods tend to predict different protein-protein interface patterns. We made use of ANOVA and variable selection to prove our result. Our proposed methods taking advantages of different single methods also got a good prediction result compared to single methods. In addition to the prediction of protein-protein interactions, this idea can be extended to other research areas such as protein structure prediction and design.
| S-EPMC5700192 | biostudies-literature
Project description:The biogeographical ancestry (BGA) of a trace or a person/skeleton refers to the component of ethnicity, constituted of biological and cultural elements, that is biologically determined. Nowadays, many individuals are interested in exploring their genealogy, and the capability to distinguish biogeographic information about population groups and subgroups via DNA analysis plays an essential role in several fields such as in forensics. In fact, for investigative and intelligence purposes, it is beneficial to inference the biogeographical origins of perpetrators of crimes or victims of unsolved cold cases when no reference profile from perpetrators or database hits for comparative purposes are available. Current approaches for biogeographical ancestry estimation using SNPs data are usually based on PCA and Structure software. The present study provides an alternative method that involves multivariate data analysis and machine learning strategies to evaluate BGA discriminating power of unknown samples using different commercial panels. Starting from 1000 Genomes project, Simons Genome Diversity Project and Human Genome Diversity Project datasets involving African, American, Asian, European and Oceania individuals, and moving towards further and more geographically restricted populations, powerful multivariate techniques such as Partial Least Squares-Discriminant Analysis (PLS-DA) and machine learning techniques such as XGBoost were employed, and their discriminating power was compared. PLS-DA method provided more robust classifications than XGBoost method, showing that the adopted approach might be an interesting tool for forensic experts to infer BGA information from the DNA profile of unknown individuals, but also highlighting that the commercial forensic panels could be inadequate to discriminate populations at intra-continental level.
| S-EPMC9148302 | biostudies-literature
Project description:Assays that are capable of detecting genome-wide chromatin interactions have produced massive amount of data and led to great understanding of the chromosomal three-dimensional (3D) structure. As technology becomes more sophisticated, higher-and-higher resolution data are being produced, going from the initial 1 Megabases (Mb) resolution to the current 10 Kilobases (Kb) or even 1 Kb resolution. The availability of genome-wide interaction data necessitates development of analytical methods to recover the underlying 3D spatial chromatin structure, but challenges abound. Most of the methods were proposed for analyzing data at low resolution (1 Mb). Their behaviors are thus unknown for higher resolution data. For such data, one of the key features is the high proportion of "0" contact counts among all available data, in other words, the excess of zeros.To address the issue of excess of zeros, in this paper, we propose a truncated Random effect EXpression (tREX) method that can handle data at various resolutions. We then assess the performance of tREX and a number of leading existing methods for recovering the underlying chromatin 3D structure. This was accomplished by creating in-silico data to mimic multiple levels of resolution and submit the methods to a "stress test". Finally, we applied tREX and the comparison methods to a Hi-C dataset for which FISH measurements are available to evaluate estimation accuracy.The proposed tREX method achieves consistently good performance in all 30 simulated settings considered. It is not only robust to resolution level and underlying parameters, but also insensitive to model misspecification. This conclusion is based on observations made in terms of 3D structure estimation accuracy and preservation of topologically associated domains. Application of the methods to the human lymphoblastoid cell line data on chromosomes 14 and 22 further substantiates the superior performance of tREX: the constructed 3D structure from tREX is consistent with the FISH measurements, and the corresponding distances predicted by tREX have higher correlation with the FISH measurements than any of the comparison methods.An open-source R-package is available at http://www.stat.osu.edu/~statgen/Software/tRex.
| S-EPMC4744395 | biostudies-literature