Identification of population-informative markers from high-density genotyping data through combined feature selection and machine learning algorithms: Application to European autochthonous and cosmopolitan pig breeds.

Large genotyping datasets, obtained from high-density single nucleotide polymorphism (SNP) arrays, developed for different livestock species, can be used to describe and differentiate breeds or populations. To identify the most discriminating genetic markers among thousands of genotyped SNPs, a few statistical approaches have been proposed. In this study, we applied the Boruta algorithm, a wrapper of the machine learning random forest algorithm, on a database of 23 European pig breeds (20 autochthonous and three cosmopolitan breeds) genotyped with a 70k SNP chip, to pre-select informative SNPs. To identify different sets of SNPs, these pre-selected markers were then ranked with random forest based on their mean decrease accuracy and mean decrease gene indexes. We evaluated the efficiency of these subsets for breed classification and the usefulness of this approach to detect candidate genes affecting breed-specific phenotypes and relevant production traits that might differ among breeds. The lowest overall classification error (2.3%) was reached with a subpanel including only 398 SNPs (ranked based on their mean decrease accuracy), with no classification error in seven breeds using up to 49 SNPs. Several SNPs of these selected subpanels were in genomic regions in which previous studies had identified signatures of selection or genes associated with morphological or production traits that distinguish the analysed breeds. Therefore, even if these approaches have not been originally designed to identify signatures of selection, the obtained results showed that they could potentially be useful for this purpose.

Comparative targeted metabolomic profiles of porcine plasma and serum.

Metabolomics has been used to characterise many biological matrices and obtain detailed pictures of biological systems based on many metabolites. Plasma and serum are two blood-derived biofluids commonly used to assess and monitor the organismal metabolism and obtain information on the physiological and health conditions of an animal. Plasma is the supernatant that is separated from the cellular components after centrifugation of the blood that is first added with an anticoagulant. Serum is obtained after centrifugation of the blood that has been coagulated. The choice of one or the other biofluid for metabolomic analyses is related to specific analytical needs and technical issues, to problems derived by the collection and preparation steps, in particular when specimens are sampled from animals involved in field studies. Thus far, most of the metabolomic studies that compared plasma and serum have been carried out in humans and very little is known on the pigs. In this study, we used a targeted metabolomic platform that can detect about 180 metabolites of five biochemical classes to compare plasma and serum profiles of samples collected from 24 pigs. To also obtain a cross-species comparative metabolomic analysis, information for human plasma and serum derived from the same platform was retrieved from previous studies. Statistical analyses included univariate and multivariate approaches aimed at identifying stable and/or differentially abundant metabolites between the two porcine biofluids. A total of 154 (∼83%) metabolites passed the initial quality control, indicating a good repeatability of the analytical platform in pigs. Discarded metabolites included aspartate and biogenic amines that were already reported to be unstable in human studies. More than 80% of the metabolites had similar profiles in both porcine biofluids (average correlation was 0.75). Concentrations were usually higher in serum than in plasma, in agreement with what was already reported in humans. The univariate analysis identified 44 metabolites that had statistically different concentrations between porcine plasma and serum, of which 28 metabolites were also confirmed by the multivariate analysis. The obtained picture described similarities and differences between these two biofluids in pigs and the related human-pig comparisons. The obtained information can be useful for the choice of one or the other matrix for the implementation of metabolomic studies in this livestock species. The results can also provide useful hints to valuing the pig as animal model, in particular when metabolite-derived physiological states are relevant.

Genome-wide association studies for 30 haematological and blood clinical-biochemical traits in Large White pigs reveal genomic regions affecting intermediate phenotypes.

Haematological and clinical-biochemical parameters are considered indicators of the physiological/health status of animals and might serve as intermediate phenotypes to link physiological aspects to production and disease resistance traits. The dissection of the genetic variability affecting these phenotypes might be useful to describe the resilience of the animals and to support the usefulness of the pig as animal model. Here, we analysed 15 haematological and 15 clinical-biochemical traits in 843 Italian Large White pigs, via three genome-wide association scan approaches (single-trait, multi-trait and Bayesian). We identified 52 quantitative trait loci (QTLs) associated with 29 out of 30 analysed blood parameters, with the most significant QTL identified on porcine chromosome 14 for basophil count. Some QTL regions harbour genes that may be the obvious candidates: QTLs for cholesterol parameters identified genes (ADCY8, APOB, ATG5, CDKAL1, PCSK5, PRL and SOX6) that are directly involved in cholesterol metabolism; other QTLs highlighted genes encoding the enzymes being measured [ALT (known also as GPT) and AST (known also as GOT)]. Moreover, the multivariate approach strengthened the association results for several candidate genes. The obtained results can contribute to define new measurable phenotypes that could be applied in breeding programs as proxies for more complex traits.

Quantifying API polymorphs in formulations using X-ray powder diffraction and multivariate standard addition method combined with net analyte signal analysis.

The direct quantification of Active Pharmaceutical Ingredients in solid formulations is a challenging open issue. A consolidated analytical technique based on X-ray Powder Diffraction is available, being the definitive test for the identification of polymorphs and crystal phases. However, its application for quantitative analysis is hindered by matrix effects: refinement methods (e.g. Rietveld method) require a complete knowledge of samples' composition, while univariate calibration methods require the matrix effect to be studied and severely suffer from the co-presence of crystalline and amorphous phases in the sample. Multivariate analysis is the only way to bypass problems affecting refinements procedures and univariate calibration. In particular, the multivariate standard addition method (SAM) is promising; however, it is straightforward only when the analytical blank (matrix devoid of analyte) is available: in that case SAM is applied by simply extrapolating the SAM model to the matrix experimental signal. In this work, the quantitative analysis of polymorphic forms of Active Pharmaceutical Ingredients based on X-ray Powder Diffraction is performed for the first time by a method based on multivariate standard addition method combined with net analyte signal procedure; it allows for reliable quantification of polymorphs of active principles in solid formulations, which are rapidly analyzed without any sample pre-treatment. Two test cases are presented: quantification of two polymorphs of piracetam in binary mixtures (forms II and III), and quantification of paracetamol (form I) in Tachifludec®.

A comparative analysis of label-free liquid chromatography-mass spectrometry liver proteomic profiles highlights metabolic differences between pig breeds.

The liver is a complex organ governing several physiological processes that define biological mechanisms affecting growth, feed efficiency and performance traits in all livestock species, including pig. Proteomics may contribute to a better understanding of the relationship between liver functions and complex production traits in pigs and to characterize this species as biomedical model. This study applied, for the first time, a label-free liquid chromatography-mass spectrometry (LC-MS) proteomic approach to compare the liver proteome profiles of two important heavy pig breeds, Italian Duroc and Italian Large White. Liver specimens were collected (after slaughtering) from performance tested pigs of these two breeds, raised in standard conditions. The label-free LC-MS method captured a total of 501 proteins of which 200 were subsequently considered in the between breeds comparison. A statistical pipeline based on the sparse Partial Least Squares Discriminant Analysis (sPLS-DA), coupled with stability and significance tests, was applied for the identification of up or down regulated proteins between breeds. This analysis revealed a total of 25 proteins clearly separating Italian Duroc and Italian Large White pigs. Among the top proteins differentiating the two breeds, 3-ketoacyl-CoA thiolase, mitochondrial (ACAA2) and histone H2B type 2-F (HIST2H2BF) were up-regulated in Italian Duroc pigs and carboxylesterase 3 (CES3) and ketohexokinase (KHK) were up-regulated in Italian Large White pigs. Fatty acid synthase (FASN), involved in fatty acid metabolism and encoded by a gene located in a QTL region for fatty acid composition, was up-regulated in Italian Large White pigs. The in silico protein interaction analysis showed that 16 of these proteins were connected in one big module. Bioinformatic functional analysis indicated that differentially expressed proteins were involved in several biological processes related to the metabolism of lipids, amino-acids, carbohydrates, cofactors and antibiotics/drugs, suggesting that these functions might distinguish Italian Duroc and Italian Large White pigs. This pilot comparative proteomic analysis of the porcine liver highlighted several biological factors that could determine the peculiar production potentials of these two heavy pig breeds, derived by their different genetic backgrounds.

A genome wide association study for backfat thickness in Italian Large White pigs highlights new regions affecting fat deposition including neuronal genes.

BACKGROUND: Carcass fatness is an important trait in most pig breeding programs. Following market requests, breeding plans for fresh pork consumption are usually designed to reduce carcass fat content and increase lean meat deposition. However, the Italian pig industry is mainly devoted to the production of Protected Designation of Origin dry cured hams: pigs are slaughtered at around 160 kg of live weight and the breeding goal aims at maintaining fat coverage, measured as backfat thickness to avoid excessive desiccation of the hams. This objective has shaped the genetic pool of Italian heavy pig breeds for a few decades. In this study we applied a selective genotyping approach within a population of ~ 12,000 performance tested Italian Large White pigs. Within this population, we selectively genotyped 304 pigs with extreme and divergent backfat thickness estimated breeding value by the Illumina PorcineSNP60 BeadChip and performed a genome wide association study to identify loci associated to this trait. RESULTS: We identified 4 single nucleotide polymorphisms with P≤5.0E-07 and additional 119 ones with 5.0E-07

Evaluating patient satisfaction through latent class factor analysis.

This paper introduces Health and Place readers interested in studying the latent concept of satisfaction to the methodology of latent variable analysis. In particular, some suitable methods for analyzing individual opinions expressed on ordinal scales are illustrated. The basic theory behind these methods is explained and a step by step description of how they should be used in practice is given. The discussion of the subject starts with the simplest methods, in which opinions are grouped into two categories: typically positive and negative. Furthermore, more complex methods are presented to deal with opinions expressed on ordinal scales (e.g., very satisfied, somewhat satisfied, and not satisfied). All methods are described by showing various results obtained through the analysis of a dataset containing patients' opinions about their satisfaction with hospital care, collected through a survey conducted after their discharge from an Italian hospital. The database was created using a questionnaire covering different aspects of satisfaction and a five-point Likert scale. This represents an example of multi-level data: patients are clustered according to the hospital ward in which they were hospitalized. Thus, some specific latent variable methods able to deal with this particular structure of the data are also described.