Multivariate regression analysis of dairy characteristics of Holstein cows

The purpose of the research is the possibility of applying the equation of multiple two-factor regression to reveal the reliable conjugacy of signs affecting the level of milk productivity of Holstein cows. Research methods. The research was carried out in CJSC BP “Aksin’ino” of Stupinskiy district of the Moscow region. Based on the IAS “SELEX”, a database was created, including a sample of 11 017 heads. Data on milk productivity were taken for 305 days of 1st, 2nd, 3rd and maximum lactation. Results. For 305 days of the 1st lactation, milk yield averaged 7909.5 kg of milk, for the 2nd – 8289.1 kg (p ≤ 0,001) and the 3rd lactation – 8446.2 kg (p ≤ 0,001). Milk yield for maximum lactation was 8964.3 kg of milk (p ≤ 0.001). The fat and protein content in cow milk between the 1st and 3rd lactation is 4.03–4.08 and 3.22–3.23 %. The multiple two-factor regression coefficient represents the response bias from 7787.81–8239.00 (1st, 2nd, 3rd lactation) to 8841.63 (maximum lactation). The scattering diagrams of the multiple regression model show that the value of the variable "milk yield" is statistically dependent on the indicators of the mass fraction of fat and the mass fraction of protein in milk. The coefficient of determination at the level of 0.997–0.998 indicates that the regression equation explains 99.7–99.8 % of the variance of the effective feature. The significance of Fischer’s F-test indicates the high reliability of the results and the absence of randomness and the presence of a pattern justified in our study. The fat and protein content had a negative relationship with milk yield (p ≤ 0,001) by lactation. High values of the relationship between milk yield and protein content (–0.518…–0.766), fat and protein content (0.626–0,784) were obtained. Scientific novelty. For the first time, studies were conducted on the correspondence of the equation model to the experimental data obtained, and the presence of the number of independent variables (mass fraction of fat and protein) included in the equation to describe the dependent variable (milk yield).

Negative Binomial factor regression with application to microbiome data analysis

The human microbiome provides essential physiological functions and helps maintain host homeostasis via the formation of intricate ecological host-microbiome relationships. While it is well established that the lifestyle of the host, dietary preferences, demographic background, and health status can influence microbial community composition and dynamics, robust generalizable associations between specific host-associated factors and specific microbial taxa have remained largely elusive. Here, we propose factor regression models that allow the estimation of structured parsimonious associations between host-related features and amplicon-derived microbial taxa. To account for the overdispersed nature of the amplicon sequencing count data, we propose Negative Binomial reduced rank regression (NB-RRR) and Negative Binomial co-sparse factor regression (NB-FAR). While NB-RRR encodes the underlying dependency among the microbial abundances as outcomes and the host-associated features as predictors through a rank-constrained coefficient matrix, NB-FAR uses a sparse singular value decomposition of the coefficient matrix. The latter approach avoids the notoriously difficult joint parameter estimation by extracting sparse unit-rank components of the coefficient matrix sequentially. To solve the non-convex optimization problems associated with these factor regression models, we present a novel iterative block-wise majorization procedure. Extensive simulation studies and an application to the microbial abundance data from the American Gut Project demonstrate the efficacy of the proposed procedure. In the American Gut Project data, we identify key factors that strongly link dietary habits and host life style to specific microbial families.

Triticale flour in bakery and rusk products

This paper researches the properties of two triticale varieties and three selection lines developed in Bashkortostan. The organoleptic indicators were typical of a standard grain; all the samples had a high protein content level. Gluten weight ratio matched the grain genotype. The falling number of the studied samples was low. Hydrothermal processing modes for triticale grain before grinding were optimized. The studied varieties and lines were found to yield 65.15%–70.18% of the flour. The conducted polynomial two-factor regression analysis proved that the flour yield factor is the grain softening period lasting 6 h. The dough deformation energy of the obtained flour samples was 67–129 units, the falling number 104–155 s indicates the low baking properties of the flour, thereby it was suggested to make brittle and crisp bakery products as crispbread and breadsticks. A breadstick formula with 60% wheat flour replaced with triticale flour was developed. The number of spore Rope bacteria increased slightly during the shelf life but did not exceed the permissible values. There have been developed formulas and methods to make crispbreads for producing crackers with triticale bran of 40% have been developed.

Investigation of the Rate of Vacuum-Conductive Drying of Wood-Particle Panels Based on Polyvinyl Alcohol

Wood composites are promising construction and finishing materials that offer the best properties of both wood and polymers. With the industry's need to use low-toxic adhesives in view, one of the ways to ensure the environmental friendliness of composites is the use of polyvinyl alcohol as a binder. In this respect, in order to streamline production, both in terms of reducing energy consumption and improving the quality of the final product, it is vital to choose the drying rate when pressing panels. The paper presents the results of a study of the process of vacuum-conductive drying of eco-friendly polyvinyl alcohol (PVA)-based wood-particle material. It was found that the composite samples with a PVA content of 40% in terms of physical and mechanical parameters (ultimate bending strength and ultimate tensile strength perpendicular to the panel face) fully comply with the established standards. The study provided the values of drying rate depending on 5 factors. The multi-factor regression analysis method was used to derive equations that determine the drying rate of polyvinyl alcohol-based wood-particle panels for two humidity intervals (above and below the cell walls humidification limit), taking into account the properties of different classes of bodies that make up the composite.