1,720,981 research outputs found
New phenotypes predictions obtained by innovative infrared spectroscopy calibrations and their genetic analysis in dairy cattle populations
Full text linkThe main objective of this thesis was to assess the infrared spectroscopy for the prediction at individual level of “new phenotypes” related to the technological properties of the cow milk, testing classic and innovative statistical approaches and evaluating the genetic parameters for a possible inclusion of the predicted traits in the selection indices as indirect selection method.
A total of 1,264 individual milk samples were used for an individual model cheese making procedure and 7 new cheese-making related traits were obtained: 3 measures of cheese yield as percentage of processed milk (%CYs; fresh cheese yield, total solids cheese yield, water retained in the curd) and 4 measures of milk nutrients retained in the curd or lost in the whey (%RECs; fat, protein, total solids and energy). The traditional milk coagulation properties (rennet coagulation time, RCT; curd firming time, k20; curd firmness at 30 and 45 min, a30 and a45 respectively ) were also measured using a Formagraph (Foss Electric A/S, Hillerød, Denmark) in a curd firmness (CF) testing time of 90 min. Using all the 360 information of the CF test recorded for each sample over the 90 min, some new modeled parameters were also obtained (modeled rennet coagulation time, RCTeq; asymptotical potential value of CF at an infinite time, CFP; curd-firming rate constant, kCF; curd-syneresis rate constant, kSR; maximum level of CF, CFmax; time at which CF attains the maximum level, tmax;). For each sample two Fourier-transform infrared (FTIR) spectra were collected with a MilkoScan FT6000 (Foss Electric, Hillerød, Denmark) over the spectral range from 5,000 to 900 wavenumber × cm-1, and averaged before data analysis. A first chemometric process was carried out, using the WinISI II software (Infrasoft International LLC, State College, PA) in which the partial least square regression (PLS) models are implemented, for the prediction of %CYs and %RECs. High prediction accuracies were found except for the fat recovery. In order to improve the prediction accuracy, Bayesian models, commonly used for genomic data, were tested and compared with PLS models.
The results have shown that for those traits that are difficult to be predicted, the Bayesian models perform better than PLS. Using an external validation procedure, the PLS was used for the prediction of %CYs and %RECs, while the BayesB model was used for the prediction of MCP and CF modeled parameters. In both cases the prediction accuracy found in validation, ranged from low to moderate. The genetic parameters of the predicted were estimated through a bivariate Bayesian analysis and linear models. Despite the low-moderate prediction accuracy in validation, the heritabilities of the predicted values were similar or higher than those of the corresponding measured values. The indirect selection of the studied traits was assessed through the genetic correlations between measured and predicted values, and the results shown that even when the coefficient of determination for the validation was moderate, the genetic correlations between predicted and measured values were always higher than the phenotypic correlations, and in the majority of cases near or higher than 90%.
The calibrations developed for the %CYs and %RECs have been used to obtain the predictions on a population data set consisting of about 200,000 spectra of individual milk samples of Holstein, Brown Swiss and Simmental dairy cows. The genetic parameters of the predicted traits were estimated and the heritability values were comparable to those of the measured traits. The genetic correlations of %CYs and %RECs with milk production and composition provide evidence that the current selection paradigm used in dairy cattle may have a limited effects on the technological parameters. Milk protein and fat content do not explain all the genetic variations of %CYs and (in particular) %RECs, thus, these traits could be directly selected to improve the cheese making aptitude of milk and its correlated economic valu
Comparison among different FT-MIR spectra treatments for the prediction of coagulation properties of individual milk of Brown Swiss
No full textComparison among different FT-MIR spectra treatments for the prediction of coagulation properties of individual milk of Brown Swiss
No full text
Interrelationships among physical and chemical traits of cheese: Explanatory latent factors and clustering of 37 categories of cheeses
Full text linkCheese presents extensive variability in physical, chemical, and sensory characteristics according to the variety of processing methods and conditions used to create it. Relationships between the many characteristics of cheeses are known for single cheese types or by comparing a few of them, but not for a large number of cheese types. This case study used the properties recorded on 1,050 different cheeses from 107 producers grouped into 37 categories to analyze and quantify the interrelationships among the chemical and physical properties of many cheese types. The 15 cheese traits considered were ripening length, weight, firmness, adhesiveness, 6 different chemical characteristics, and 5 different color traits. As the 105 correlations between the 15 cheese traits were highly variable, a multivariate analysis was carried out. Four latent explanatory factors were extracted, representing 86% of the covariance matrix: the first factor (38% of covariance) was named Solids because it is mainly linked positively to fat, protein, water-soluble nitrogen, ash, firmness, adhesiveness, and ripening length, and negatively to moisture and lightness; the second factor (24%) was named Hue because it is linked positively to redness/blueness, yellowness/greenness, and chroma, and negatively to hue; the third factor (17%) was named Size because it is linked positively to weight, ripening length, firmness, and protein; and the fourth factor (7%) was named Basicity because it is linked positively to pH. The 37 cheese categories were grouped into 8 clusters and described using the latent factors: the Grana Padano cluster (characterized mainly by high Size scores); hard mountain cheeses (mainly high Solids scores); very soft cheeses (low Solids scores); blue cheeses (high Basicity scores), yellowish cheeses (high Hue scores), and 3 other clusters (soft cheeses, pasta filata and treated rind, and firm mountain cheeses) according to specific combinations of intermediate latent factors and cheese traits. In this case study, the high variability and interdependence of 15 major cheese traits can be substantially explained by only 4 latent factors, allowing us to identify and characterize 8 cheese type clusters
Bayesian regression models outperform partial least squares methods for predicting milk components and technological properties using infrared spectral data
Full text linkThe aim of this study was to assess the performance of Bayesian models commonly used for genomic selection to predict "difficult-to-predict" dairy traits, such as milk fatty acid (FA) expressed as percentage of total fatty acids, and technological properties, such as fresh cheese yield and protein recovery, using Fourier-transform infrared (FTIR) spectral data. Our main hypothesis was that Bayesian models that can estimate shrinkage and perform variable selection may improve our ability to predict FA traits and technological traits above and beyond what can be achieved using the current calibration models (e.g., partial least squares, PLS). To this end, we assessed a series of Bayesian methods and compared their prediction performance with that of PLS. The comparison between models was done using the same sets of data (i.e., same samples, same variability, same spectral treatment) for each trait. Data consisted of 1,264 individual milk samples collected from Brown Swiss cows for which gas chromatographic FA composition, milk coagulation properties, and cheese-yield traits were available. For each sample, 2 spectra in the infrared region from 5,011 to 925cm(-1) were available and averaged before data analysis. Three Bayesian models: Bayesian ridge regression (Bayes RR), Bayes A, and Bayes B, and 2 reference models: PLS and modified PLS (MPLS) procedures, were used to calibrate equations for each of the traits. The Bayesian models used were implemented in the R package BGLR (http://cran.r-project.org/web/packages/BGLR/index.html), whereas the PLS and MPLS were those implemented in the WinISI II software (Infrasoft International LLC, State College, PA). Prediction accuracy was estimated for each trait and model using 25 replicates of a training-testing validation procedure. Compared with PLS, which is currently the most widely used calibration method, MPLS and the 3 Bayesian methods showed significantly greater prediction accuracy. Accuracy increased in moving from calibration to external validation methods, and in moving from PLS and MPLS to Bayesian methods, particularly Bayes A and Bayes B. The maximum R(2) value of validation was obtained with Bayes B and Bayes A. For the FA, C10:0 (% of each FA on total FA basis) had the highest R(2) (0.75, achieved with Bayes A and Bayes B), and among the technological traits, fresh cheese yield R(2) of 0.82 (achieved with Bayes B). These 2 methods have proven to be useful instruments in shrinking and selecting very informative wavelengths and inferring the structure and functions of the analyzed traits. We conclude that Bayesian models are powerful tools for deriving calibration equations, and, importantly, these equations can be easily developed using existing open-source software. As part of our study, we provide scripts based on the open source R software BGLR, which can be used to train customized prediction equations for other traits or populations
Going Beyond Counting First Authors in Author Co-citation Analysis
Full text linkThe present study examines one of the fundamental aspects of author co-citation analysis (ACA) - the way co-citation
counts are defined. Co-citation counting provides the data on which all subsequent statistical analyses and mappings
are based, and we compare ACA results based on two different types of co-citation counting - the traditional type that
only counts the first one among a cited work's authors on the one hand and a non-traditional type that takes into
account the first 5 authors of a cited work on the other hand. Results indicate that the picture produced through this non-traditional author co-citation counting contains more coherent author groups and is therefore considerably clearer. However, this picture represents fewer specialties in the research field being studied than that produced through the traditional first-author co-citation counting when the same number of top-ranked authors is selected and analyzed. Reasons for these effects are discussed
- …
