General Elements of Genomic Selection and Statistical Learning

Nowadays, huge data quantities are collected and analyzed for delivering deep insights into biological processes and human behavior. This chapter assesses the use of big data for prediction and estimation through statistical machine learning and its applications in agriculture and genetics in general, and specifically, for genome-based prediction and selection. First, we point out the importance of data and how the use of data is reshaping our way of living. We also provide the key elements of genomic selection and its potential for plant improvement. In addition, we analyze elements of modeling with machine learning methods applied to genomic selection and stress their importance as a predictive methodology. Two cultures of model building are analyzed and discussed: prediction and inference; by understanding modeling building, researchers will be able to select the best model/method for each circumstance. Within this context, we explain the differences between nonparametric models (predictors are constructed according to information derived from data) and parametric models (all the predictors take predetermined forms with the response) as well their type of effects: fixed, random, and mixed. Basic elements of linear algebra are provided to facilitate understanding the contents of the book. This chapter also contains examples of the different types of data using supervised, unsupervised, and semi-supervised learning methods.

Parametric and Nonparametric Population Pharmacokinetic Models to Assess Probability of Target Attainment of Imipenem Concentrations in Critically Ill Patients

Population pharmacokinetic modeling and simulation (M&S) are used to improve antibiotic dosing. Little is known about the differences in parametric and nonparametric M&S. Our objectives were to compare (1) the external validation of parametric and nonparametric models of imipenem in critically ill patients and (2) the probability of target attainment (PTA) calculations using simulations of both models. The M&S software used was NONMEM 7.2 (parametric) and Pmetrics 1.5.2 (nonparametric). The external predictive performance of both models was adequate for eGFRs ≥ 78 mL/min but insufficient for lower eGFRs, indicating that the models (developed using a population with eGFR ≥ 60 mL/min) could not be extrapolated to lower eGFRs. Simulations were performed for three dosing regimens and three eGFRs (90, 120, 150 mL/min). Fifty percent of the PTA results were similar for both models, while for the other 50% the nonparametric model resulted in lower MICs. This was explained by a higher estimated between-subject variability of the nonparametric model. Simulations indicated that 1000 mg q6h is suitable to reach MICs of 2 mg/L for eGFRs of 90–120 mL/min. For MICs of 4 mg/L and for higher eGFRs, dosing recommendations are missing due to largely different PTA values per model. The consequences of the different modeling approaches in clinical practice should be further investigated.

The ENDS of assumptions; an online tool for the Epistemic Nonparametric Drug-response Scoring

MotivationThe drugs sensitivity analysis is often elucidated from drug dose-response curves. These curves capture the degree of cell viability (or inhibition) over a range of induced drugs, often with parametric assumptions that are rarely validated.ResultsWe present a class of nonparametric models for the curve fitting and scoring of drug dose-responses. To allow a more objective representation of the drug sensitivity, these epistemic models devoid of any parametric assumptions attached to the linear fit, allow the parallel indexing such as IC50 and AUC. Specifically, three nonparametric models including Spline, Monotonic, and Bayesian (npS, npM, npB) and the parametric Logistic (pL) are implemented. Other indices including Maximum Effective Dose (MED) and Drug-response Span Gradient (DSG) pertinent to the npS are also provided to facilitate the interpretation of the fit. The collection of these models are implemented in an online app, standing as useful resource for drug dose-response curve fitting and analysis.AvailabilityThe ENDS is freely available online at https://irscope.shinyapps.io/ENDS/ and source codes can be obtained from https://github.com/AmiryousefiLab/ENDS.Supplementary informationSupplementary data are available at Bioinformatics and https://irscope.shinyapps.io/ENDS/[email protected]; [email protected] conceived the study and developed the models, AA and BW adopted and implemented the methods, JT provided the funding, AA, BW, MJ, and JT wrote the paper.

Nonparametric bounds on treatment effects with imperfect instruments

This paper extends the identification results in Nevo and Rosen (2012) to nonparametric models. We derive nonparametric bounds on the average treatment effect when an imperfect instrument is available. As in Nevo and Rosen (2012), we assume that the correlation between the imperfect instrument and the unobserved latent variables has the same sign as the correlation between the endogenous variable and the latent variables. We show that the monotone treatment selection and monotone instrumental variable restrictions, introduced by Manski and Pepper (2000, 2009), jointly imply this assumption. Moreover, we show how the monotone treatment response assumption can help tighten the bounds. The identified set can be written in the form of intersection bounds, which is more conducive to inference. We illustrate our methodology using the National Longitudinal Survey of Young Men data to estimate returns to schooling.

Targeting millers to improve rice marketing in Uganda

PurposeImproving milling quality is expected to improve the quality of domestic rice and hence the competitiveness of Uganda's rice industry. Therefore, this study aims to assess the determinants of four aspects of milling, namely, choice of milling technology, millers' perceptions of the importance of paddy quality attributes, milling return and milling capacity.Design/methodology/approachMultinomial logit, semi-nonparametric extended ordered probit, linear regression and additive nonparametric models are applied to cross-sectional data obtained from a sample of 196 rice millers.FindingsPhysical, economic, institutional, technological and sociodemographic factors are found to be important determinants of the four aspects of milling. Physical factors include the distance of the mill from major town and availability of storage space at the milling premises, while economic factors include milling charge and backward integration of miller into paddy production. Contracting and use of a single-pass mill are important institutional and technological factors, respectively, and miller's household size, age, gender and education are the key sociodemographic variables.Originality/valueThe study's originality lies in its scope, especially in terms of its breadth. Without compromising the needed analytical rigor, it focuses on four aspects of milling that are critical to improving the marketing of Uganda's rice. In doing so, it provides a holistic understanding of this segment of the value chain and offers specific recommendations for improving the marketing of Uganda's rice.

Identification of a Set of Variables for the Classification of Páramo Soils Using a Nonparametric Model, Remote Sensing, and Organic Carbon

Páramo ecosystems harbor important biodiversity and provide essential environmental services such as water regulation and carbon sequestration. Unfortunately, the scarcity of information on their land uses makes it difficult to generate sustainable strategies for their conservation. The purpose of this study is to develop a methodology to easily monitor and document the conservation status, degradation rates, and land use changes in the páramo. We analyzed the performance of two nonparametric models (the CART decision tree, CDT, and multivariate adaptive regression curves, MARS) in the páramos of the Chambo sub-basin (Ecuador). We used three types of attributes: digital elevation model (DEM), land use cover (Sentinel 2), and organic carbon content (Global Soil Organic Carbon Map data, GSOC) and a categorical variable, land use. We obtained a set of selected variables which perform well with both models, and which let us monitor the land uses of the páramos. Comparing our results with the last report of the Ecuadorian Ministry of Environment (2012), we found that 9% of the páramo has been lost in the last 8 years.