Home About us Contact
|
Home About us Contact | ||
|
|
||
Cross-validation Analysis (cross-validation + analysis)
Selected AbstractsSpatial prediction of river channel topography by krigingEARTH SURFACE PROCESSES AND LANDFORMS, Issue 6 2008Carl J. Legleiter Abstract Topographic information is fundamental to geomorphic inquiry, and spatial prediction of bed elevation from irregular survey data is an important component of many reach-scale studies. Kriging is a geostatistical technique for obtaining these predictions along with measures of their reliability, and this paper outlines a specialized framework intended for application to river channels. Our modular approach includes an algorithm for transforming the coordinates of data and prediction locations to a channel-centered coordinate system, several different methods of representing the trend component of topographic variation and search strategies that incorporate geomorphic information to determine which survey data are used to make a prediction at a specific location. For example, a relationship between curvature and the lateral position of maximum depth can be used to include cross-sectional asymmetry in a two-dimensional trend surface model, and topographic breaklines can be used to restrict which data are retained in a local neighborhood around each prediction location. Using survey data from a restored gravel-bed river, we demonstrate how transformation to the channel-centered coordinate system facilitates interpretation of the variogram, a statistical model of reach-scale spatial structure used in kriging, and how the choice of a trend model affects the variogram of the residuals from that trend. Similarly, we show how decomposing kriging predictions into their trend and residual components can yield useful information on channel morphology. Cross-validation analyses involving different data configurations and kriging variants indicate that kriging is quite robust and that survey density is the primary control on the accuracy of bed elevation predictions. The root mean-square error of these predictions is directly proportional to the spacing between surveyed cross-sections, even in a reconfigured channel with a relatively simple morphology; sophisticated methods of spatial prediction are no substitute for field data. Copyright © 2007 John Wiley & Sons, Ltd. [source] Environmental regulation and modelling of cassava canopy conductance under drying root-zone soil waterMETEOROLOGICAL APPLICATIONS, Issue 3 2007Philip G. Oguntunde Abstract Sap flow was measured, with Granier-type sensors, in a crop of field-grown water-stressed cassava (Manihot esculenta Crantz) in Ghana, West Africa. The main objective of this study was to examine the environmental control of canopy conductance (gc) with a view to modelling the stomatal control of water transport under water-stressed condition. Weather variables measured concurrently with sap flow were: air temperature (Ta), relative humidity (RH), wind speed (u) and solar radiation (Rs). Relationship between canopy conductance (gc) and vapour pressure deficit (D,) was curvilinear while no specific pattern was observed with Rs. Average diurnal gc decreased from 3.0 ± 0.6 to 0.7 ± 0.4 mm s,1 between 0730 and 2000 h local time ( = GMT) each day. A Jarvis-type model, based on a set of environmental control functions, was parameterized for the cassava crop in this study. Model results demonstrated that gc was estimated with a high degree of accuracy based on Rs, Ta, and D, (r2 = 0.92;F = 809.2;P < 0.0001). D, explained about 90% (F = 2129.7;P < 0.0001) of the variations observed in gc, whereas both Rs and Ta contributed about 2% of the explained variance in gc. The aerodynamic conductance (ga) was very high compared to gc, leading to a daily average ratio ga/gc > 100 and a decoupling factor < 0.1. Cross-validation analysis revealed a consistent good performance (r2 > 0.85) of the gc model with D, as the only independent environmental variable. Copyright © 2007 Royal Meteorological Society [source] Validation of whole genome linkage-linkage disequilibrium and association results, and identification of markers to predict genetic merit for twinningANIMAL GENETICS, Issue 4 2010C. D. Bierman Summary A previous genome-wide search with a moderate density 10K marker set identified many marker associations with twinning rate, either through single-marker analysis or combined linkage-linkage disequilibrium (LLD; haplotype) analysis. The objective of the current study was to validate putative marker associations using an independent set of phenotypic data. Holstein bulls (n = 921) from 100 paternal half-sib families were genotyped. Twinning rate predicted transmitting abilities were calculated using calving records from 1994 to 1998 (Data I) and 1999 to 2006 (Data II), and the underlying liability scores from threshold model analysis were used as the trait in marker association analyses. The previous analysis used 201 bulls with daughter records in Data I. In the current analysis, this was increased to 434, providing a revised estimate of effect and significance. Bulls with daughter records in Data II totaled 851, and analysis of this data provided the validation of results from analysis of Data I. Single nucleotide polymorphisms (SNPs) were selected to validate previously significant single-marker associations and LLD results. Bulls were genotyped for a total of 306 markers. Nine of 13 LLD regions located on chromosomes 1, 2, 3, 6, 9, 22, 23(2) and 26 were validated, showing significant results for both Data I and II. Association analysis revealed 55 of 174 markers validated, equating to a single-marker validation rate of 31%. Stepwise backward elimination and cross-validation analyses identified 18 SNPs for use in a final reduced marker panel explaining 34% of the genetic variation, and to allow prediction of genetic merit for twinning rate. [source] NIR, DSC, and FTIR as quantitative methods for compositional analysis of blends of polymers obtained from recycled mixed plastic wastePOLYMER ENGINEERING & SCIENCE, Issue 9 2001Walker Camacho Methods for the determination of the composition of two binary blends in mixtures of recycled polymeric materials were analyzed and compared. Recycled polypropylene/polyethylene (PP/HDPE) and recycled poly(acryl-butadiene-styrene) and polypropylene(ABS/PP) were used to develop and validate the methods. Diffuse reflectance near infrared spectroscopy (NIRS) offers high sensitivity and ease of operation and a possibility to perform multivariate data analysis. In comparison, differential scanning calorimetry (DSC) and Mid-IR, which are commonly used for this purpose require certain sample preparation and are indeed time consuming. In addition, the low sensitivity of these two methods to concentrations lower than 1% wt makes their application in quality control of recycled polymers inappropriate. NIR can be used for estimating the composition of the recyclate on-line in only a few seconds, no sample preparation is required, and high precision is achieved. We obtained a root mean square error of prediction (RMSEP) equal to 0.21% wt in the interval from 0-15% wt of PP in HDPE and a RMSEP equal to 0.91% wt in the interval 0-100%. For blends of PP/ABS a RMSEP of 0.74% wt in the range 0-100% and 0.32% wt in the range 0-15% wt PP was calculated. Most of the variation in the spectral data with respect to the polymer blend composition for all the studied blends were explained by two principal components (PC). The optimal number of factors (PC) was determined by cross-validation analysis. [source] Prediction of reversibly oxidized protein cysteine thiols using protein structure propertiesPROTEIN SCIENCE, Issue 3 2008Ricardo Sanchez Abstract Protein cysteine thiols can be divided into four groups based on their reactivities: those that form permanent structural disulfide bonds, those that coordinate with metals, those that remain in the reduced state, and those that are susceptible to reversible oxidation. Physicochemical parameters of oxidation-susceptible protein thiols were organized into a database named the Balanced Oxidation Susceptible Cysteine Thiol Database (BALOSCTdb). BALOSCTdb contains 161 cysteine thiols that undergo reversible oxidation and 161 cysteine thiols that are not susceptible to oxidation. Each cysteine was represented by a set of 12 parameters, one of which was a label (1/0) to indicate whether its thiol moiety is susceptible to oxidation. A computer program (the C4.5 decision tree classifier re-implemented as the J48 classifier) segregated cysteines into oxidation-susceptible and oxidation-non-susceptible classes. The classifier selected three parameters critical for prediction of thiol oxidation susceptibility: (1) distance to the nearest cysteine sulfur atom, (2) solvent accessibility, and (3) pKa. The classifier was optimized to correctly predict 136 of the 161 cysteine thiols susceptible to oxidation. Leave-one-out cross-validation analysis showed that the percent of correctly classified cysteines was 80.1% and that 16.1% of the oxidation-susceptible cysteine thiols were incorrectly classified. The algorithm developed from these parameters, named the Cysteine Oxidation Prediction Algorithm (COPA), is presented here. COPA prediction of oxidation-susceptible sites can be utilized to locate protein cysteines susceptible to redox-mediated regulation and identify possible enzyme catalytic sites with reactive cysteine thiols. [source] | ||||||||||