In silico prediction of blood cholesterol levels from genotype data.
Ontology highlight
ABSTRACT:
SUBMITTER: Reggiani F
PROVIDER: S-EPMC7010235 | biostudies-literature | 2020
REPOSITORIES: biostudies-literature
Publications
In silico prediction of blood cholesterol levels from genotype data.
Reggiani Francesco F Carraro Marco M Belligoli Anna A Sanna Marta M Dal Prà Chiara C Favaretto Francesca F Ferrari Carlo C Vettor Roberto R Tosatto Silvio C E SCE
PloS one 20200210 2
In this work we present a framework for blood cholesterol levels prediction from genotype data. The predictor is based on an algorithm for cholesterol metabolism simulation available in literature, implemented and optimized by our group in the R language. The main weakness of the former simulation algorithm was the need of experimental data to simulate mutations in genes altering the cholesterol metabolism. This caveat strongly limited the application of the model in the clinical practice. In th ...[more]
Similar Datasets
Project description:Least absolute shrinkage and selection operator (LASSO) regression is often applied to select the most promising set of single nucleotide polymorphisms (SNPs) associated with a molecular phenotype of interest. While the penalization parameter λ restricts the number of selected SNPs and the potential model overfitting, the least-squares loss function of standard LASSO regression translates into a strong dependence of statistical results on a small number of individuals with phenotypes or genotypes divergent from the majority of the study population-typically comprised of outliers and high-leverage observations. Robust methods have been developed to constrain the influence of divergent observations and generate statistical results that apply to the bulk of study data, but they have rarely been applied to genetic association studies. In this article, we review, for newcomers to the field of robust statistics, a novel version of standard LASSO that utilizes the Huber loss function. We conduct comprehensive simulations and analyze real protein, metabolite, mRNA expression and genotype data to compare the stability of penalization, the cross-iteration concordance of the model, the false-positive and true-positive rates and the prediction accuracy of standard and robust Huber-LASSO. Although the two methods showed controlled false-positive rates ≤2.1% and similar true-positive rates, robust Huber-LASSO outperformed standard LASSO in the accuracy of predicted protein, metabolite and gene expression levels using individual SNP data. The conducted simulations and real-data analyses show that robust Huber-LASSO represents a valuable alternative to standard LASSO in genetic studies of molecular phenotypes.
| S-EPMC8293825 | biostudies-literature
Project description:Predicting disease phenotypes from genotypes is a key challenge in medical applications in the postgenomic era. Large training datasets of patients that have been both genotyped and phenotyped are the key requisite when aiming for high prediction accuracy. With current genotyping projects producing genetic data for hundreds of thousands of patients, large-scale phenotyping has become the bottleneck in disease phenotype prediction.Here we present an approach for imputing missing disease phenotypes given the genotype of a patient. Our approach is based on co-training, which predicts the phenotype of unlabeled patients based on a second class of information, e.g. clinical health record information. Augmenting training datasets by this type of in silico phenotyping can lead to significant improvements in prediction accuracy. We demonstrate this on a dataset of patients with two diagnostic types of migraine, termed migraine with aura and migraine without aura, from the International Headache Genetics Consortium.Imputing missing disease phenotypes for patients via co-training leads to larger training datasets and improved prediction accuracy in phenotype prediction.The code can be obtained at: http://www.bsse.ethz.ch/mlcb/research/bioinformatics-and-computational-biology/co-training.html
| S-EPMC4765855 | biostudies-literature
Project description:Tens of thousands of RNA-sequencing experiments comprising hundreds of thousands of individual samples have now been performed. These data represent a broad range of experimental conditions, sequencing technologies, and hypotheses under study. The Recount project has aggregated and uniformly processed hundreds of thousands of publicly available RNA-seq samples. Most of these samples only include RNA expression measurements; genotype data for these same samples would enable a wide range of analyses including variant prioritization, eQTL analysis, and studies of allele specific expression. Here, we developed a statistical model based on the existing reference and alternative read counts from the RNA-seq experiments available through Recount3 to predict genotypes at autosomal biallelic loci in coding regions. We demonstrate the accuracy of our model using large-scale studies that measured both gene expression and genotype genome-wide. We show that our predictive model is highly accurate with 99.5% overall accuracy, 99.6% major allele accuracy, and 90.4% minor allele accuracy. Our model is robust to tissue and study effects, provided the coverage is high enough. We applied this model to genotype all the samples in Recount 3 and provide the largest ready-to-use expression repository containing genotype information. We illustrate that the predicted genotype from RNA-seq data is sufficient to unravel the underlying population structure of samples in Recount3 using Principal Component Analysis.
| S-EPMC10979922 | biostudies-literature
Project description:BackgroundHIV/AIDS is a serious threat to public health. The emergence of drug resistance mutations diminishes the effectiveness of drug therapy for HIV/AIDS. Developing a computational prediction of drug resistance phenotype will enable efficient and timely selection of the best treatment regimens.ResultsA unified encoding of protein sequence and structure was used as the feature vector for predicting phenotypic resistance from genotype data. Two machine learning algorithms, Random Forest and K-nearest neighbor, were used. The prediction accuracies were examined by five-fold cross-validation on the genotype-phenotype datasets. A supervised machine learning approach for automatic prediction of drug resistance was developed to handle genotype-phenotype datasets of HIV protease (PR) and reverse transcriptase (RT). It predicts the drug resistance phenotype and its relative severity from a query sequence. The accuracy of the classification was higher than 0.973 for eight PR inhibitors and 0.986 for ten RT inhibitors, respectively. The overall cross-validated regression R(2)-values for the severity of drug resistance were 0.772-0.953 for 8 PR inhibitors and 0.773-0.995 for 10 RT inhibitors.ConclusionsMachine learning using a unified encoding of sequence and protein structure as a feature vector provides an accurate prediction of drug resistance from genotype data. A practical webserver for clinicians has been implemented.
| S-EPMC5009519 | biostudies-literature
Project description:The added value of incorporating information from repeated blood pressure and cholesterol measurements to predict cardiovascular disease (CVD) risk has not been rigorously assessed. We used data on 191,445 adults from the Emerging Risk Factors Collaboration (38 cohorts from 17 countries with data encompassing 1962-2014) with more than 1 million measurements of systolic blood pressure, total cholesterol, and high-density lipoprotein cholesterol. Over a median 12 years of follow-up, 21,170 CVD events occurred. Risk prediction models using cumulative mean values of repeated measurements and summary measures from longitudinal modeling of the repeated measurements were compared with models using measurements from a single time point. Risk discrimination (C-index) and net reclassification were calculated, and changes in C-indices were meta-analyzed across studies. Compared with the single-time-point model, the cumulative means and longitudinal models increased the C-index by 0.0040 (95% confidence interval (CI): 0.0023, 0.0057) and 0.0023 (95% CI: 0.0005, 0.0042), respectively. Reclassification was also improved in both models; compared with the single-time-point model, overall net reclassification improvements were 0.0369 (95% CI: 0.0303, 0.0436) for the cumulative-means model and 0.0177 (95% CI: 0.0110, 0.0243) for the longitudinal model. In conclusion, incorporating repeated measurements of blood pressure and cholesterol into CVD risk prediction models slightly improves risk prediction.
| S-EPMC5860526 | biostudies-literature
Project description:BACKGROUND: Established risk models for the prediction of cardiovascular disease (CVD) include blood pressure, smoking and cholesterol parameters. The use of total cholesterol for CVD risk prediction has been questioned, particularly for primary prevention. We evaluated whether glucose could be used instead of total cholesterol for prediction of fatal CVD using data with long follow-up. METHODS: We followed-up 6,095 men and women aged ?16 years who participated 1977-79 in a community based health study and were anonymously linked with the Swiss National Cohort until the end of 2008. During follow-up, 727 participants died of CVD. Based on the ESC SCORE methodology (Weibull regression), we used age, sex, blood pressure, smoking, and fasting glucose or total cholesterol. The mean Brier score (BS), area under the receiver-operating characteristic curve (AUC) and integrated discrimination improvement (IDI) were used for model comparison. We validated our models internally using cross-validation and externally using another data set. RESULTS: In our models, the p-value of total cholesterol was 0.046, that of glucose was p?<?0.001. The model with glucose had a slightly better predictive capacity (BS: 2216x10-5 vs. 2232x10-5; AUC: 0.9181 vs. 0.9169, IDI: 0.009 with p-value 0.026) and could well discriminate the overall risk of persons with high and low concentrations. The external validation confirmed these findings. CONCLUSIONS: Our study suggests that instead of total cholesterol glucose can be used in models predicting overall CVD mortality risk.
| S-EPMC3599220 | biostudies-other
Project description:Computational identification of blood-secretory proteins, especially proteins with differentially expressed genes in diseased tissues, can provide highly useful information in linking transcriptomic data to proteomic studies for targeted disease biomarker discovery in serum.A new algorithm for prediction of blood-secretory proteins is presented using an information-retrieval technique, called manifold ranking. On a dataset containing 305 known blood-secretory human proteins and a large number of other proteins that are either not blood-secretory or unknown, the new method performs better than the previous published method, measured in terms of the area under the recall-precision curve (AUC). A key advantage of the presented method is that it does not explicitly require a negative training set, which could often be noisy or difficult to derive for most biological problems, hence making our method more applicable than classification-based data mining methods in general biological studies.We believe that our program will prove to be very useful to biomedical researchers who are interested in finding serum markers, especially when they have candidate proteins derived through transcriptomic or proteomic analyses of diseased tissues. A computer program is developed for prediction of blood-secretory proteins based on manifold ranking, which is accessible at our website http://csbl.bmb.uga.edu/publications/materials/qiliu/blood_secretory_protein.html.
| S-EPMC2877692 | biostudies-literature
Project description:BACKGROUND: Genes implicated in common complex disorders such as obesity, type 2 diabetes mellitus (T2DM) or cardiovascular diseases are not disease specific, since clinically related disorders also share genetic components. Cysteine protease Calpain 10 (CAPN10) has been associated with T2DM, hypertension, hypercholesterolemia, increased body mass index (BMI) and polycystic ovary syndrome (PCOS), a reproductive disorder of women in which isunlin resistance seems to play a pathogenic role. The calpain 5 gene (CAPN5) encodes a protein homologue of CAPN10. CAPN5 has been previously associated with PCOS by our group. In this new study, we have analysed the association of four CAPN5 gene variants(rs948976A>G, rs4945140G>A, rs2233546C>T and rs2233549G>A) with several cardiovascular risk factors related to metabolic syndrome in general population. METHODS: Anthropometric measurements, blood pressure, insulin, glucose and lipid profiles were determined in 606 individuals randomly chosen from a cross-sectional population-based epidemiological survey in the province of Segovia in Central Spain (Castille), recruited to investigate the prevalence of anthropometric and physiological parameters related to obesity and other components of the metabolic syndrome. Genotypes at the four polymorphic loci in CAPN5 gene were detected by polymerase chain reaction (PCR). RESULTS: Genotype association analysis was significant for BMI (p < or = 0.041), diastolic blood pressure (p = 0.015) and HDL-cholesterol levels (p = 0.025). Different CAPN5 haplotypes were also associated with diastolic blood pressure (DBP) (0.0005 < or = p < or = 0.006) and total cholesterol levels (0.001 < or = p < or = 0.029). In addition, the AACA haplotype, over-represented in obese individuals, is also more frequent in individuals with metabolic syndrome defined by ATPIII criteria (p = 0.029). CONCLUSION: As its homologue CAPN10, CAPN5 seems to influence traits related to increased risk for cardiovascular diseases. Our results also may suggest CAPN5 as a candidate gene for metabolic syndrome.
| S-EPMC1783645 | biostudies-literature
Project description:Background The role of bacteria on the onset of cardiovascular disease has been suggested. Reciprocally, increased intestinal bacterial translocation and bloodstream infection are common comorbidities associated with heart failure and myocardial infarction (MI). In this context, the aim of this study was to analyze the blood microbiome in patients shortly after acute myocardial infarction. Methods and Results We carried out a case control study comparing 103 patients at high cardiovascular risk but free of coronary disease and 99 patients who had an MI. The blood microbiome was analyzed both quantitatively by 16S quantitative polymerase chain reaction and qualitatively by 16S targeted metagenomic sequencing specifically optimized for blood samples. A significant increase in blood bacterial 16S rDNA concentration was observed in patients admitted for MI. This increase in blood bacterial DNA concentration was independent of post-MI left ventricular function and was more marked in patients with low-density lipoprotein cholesterol ?1 g/L. In addition, differences in the proportion of numerous bacterial taxa in blood were significantly modified with the onset of MI, thus defining a blood microbiota signature of MI. Among the bacterial taxa whose proportions are decreased in patients with MI, at least 6 are known to include species able to metabolize cholesterol. Conclusions These results could provide the basis for the identification of blood microbiome-based biomarkers for the stratification of MI patients. Furthermore, these findings should provide insight into the mechanism underlying the negative correlation reported between low-density lipoprotein cholesterol concentration and the prognosis at the acute onset of MI and mortality. Clinical Trial Registration URL: http://www.clinicaltrials.gov. Unique identifier: NCT02405468.
| S-EPMC6806051 | biostudies-literature
Project description:BackgroundTarget identification and validation is a pressing challenge in the pharmaceutical industry, with many of the programmes that fail for efficacy reasons showing poor association between the drug target and the disease. Computational prediction of successful targets could have a considerable impact on attrition rates in the drug discovery pipeline by significantly reducing the initial search space. Here, we explore whether gene-disease association data from the Open Targets platform is sufficient to predict therapeutic targets that are actively being pursued by pharmaceutical companies or are already on the market.MethodsTo test our hypothesis, we train four different classifiers (a random forest, a support vector machine, a neural network and a gradient boosting machine) on partially labelled data and evaluate their performance using nested cross-validation and testing on an independent set. We then select the best performing model and use it to make predictions on more than 15,000 genes. Finally, we validate our predictions by mining the scientific literature for proposed therapeutic targets.ResultsWe observe that the data types with the best predictive power are animal models showing a disease-relevant phenotype, differential expression in diseased tissue and genetic association with the disease under investigation. On a test set, the neural network classifier achieves over 71% accuracy with an AUC of 0.76 when predicting therapeutic targets in a semi-supervised learning setting. We use this model to gain insights into current and failed programmes and to predict 1431 novel targets, of which a highly significant proportion has been independently proposed in the literature.ConclusionsOur in silico approach shows that data linking genes and diseases is sufficient to predict novel therapeutic targets effectively and confirms that this type of evidence is essential for formulating or strengthening hypotheses in the target discovery process. Ultimately, more rapid and automated target prioritisation holds the potential to reduce both the costs and the development times associated with bringing new medicines to patients.
| S-EPMC5576250 | biostudies-literature