Dataset Information

A machine learning approach using partitioning around medoids clustering and random forest classification to model groups of farms in regard to production parameters and bulk tank milk antibody status of two major internal parasites in dairy cows

ABSTRACT: Fasciola hepatica and Ostertagia ostertagi are internal parasites of cattle compromising physiology, productivity, and well-being. Parasites are complex in their effect on hosts, sometimes making it difficult to identify clear directions of associations between infection and production parameters. Therefore, unsupervised approaches not assuming a structure reduce the risk of introducing bias to the analysis. They may provide insights which cannot be obtained with conventional, supervised methodology. An unsupervised, exploratory cluster analysis approach using the k–mode algorithm and partitioning around medoids detected two distinct clusters in a cross-sectional data set of milk yield, milk fat content, milk protein content as well as F. hepatica or O. ostertagi bulk tank milk antibody status from 606 dairy farms in three structurally different dairying regions in Germany. Parasite–positive farms grouped together with their respective production parameters to form separate clusters. A random forests algorithm characterised clusters with regard to external variables. Across all study regions, co–infections with F. hepatica or O. ostertagi, respectively, farming type, and pasture access appeared to be the most important factors discriminating clusters (i.e. farms). Furthermore, farm level lameness prevalence, herd size, BCS, stage of lactation, and somatic cell count were relevant criteria distinguishing clusters. This study is among the first to apply a cluster analysis approach in this context and potentially the first to implement a k–medoids algorithm and partitioning around medoids in the veterinary field. The results demonstrated that biologically relevant patterns of parasite status and milk parameters exist between farms positive for F. hepatica or O. ostertagi, respectively, and negative farms. Moreover, the machine learning approach confirmed results of previous work and shed further light on the complex setting of associations a between parasitic diseases, milk yield and milk constituents, and management practices.

SUBMITTER: Oehm A

PROVIDER: S-EPMC9273072 | biostudies-literature |

REPOSITORIES: biostudies-literature

ACCESS DATA

Similar Datasets

Project description:BackgroundPartitioning around medoids (PAM) is one of the most widely used and successful clustering method in many fields. One of its key advantages is that it only requires a distance or a dissimilarity between the individuals, and the fact that cluster centers are actual points in the data set means they can be taken as reliable representatives of their classes. However, its wider application is hampered by the large amount of memory needed to store the distance matrix (quadratic on the number of individuals) and also by the high computational cost of computing such distance matrix and, less importantly, by the cost of the clustering algorithm itself.ResultsTherefore, new software has been provided that addresses these issues. This software, provided under GPL license and usable as either an R package or a C++ library, calculates in parallel the distance matrix for different distances/dissimilarities ([Formula: see text], [Formula: see text], Pearson, cosine and weighted Euclidean) and also implements a parallel fast version of PAM (FASTPAM1) using any data type to reduce memory usage. Moreover, the parallel implementation uses all the cores available in modern computers which greatly reduces the execution time. Besides its general application, the software is especially useful for processing data of single cell experiments. It has been tested in problems including clustering of single cell experiments with up to 289,000 cells with the expression of about 29,000 genes per cell.ConclusionsComparisons with other current packages in terms of execution time have been made. The method greatly outperforms the available R packages for distance matrix calculation and also improves the packages that implement the PAM itself. The software is available as an R package at https://CRAN.R-project.org/package=scellpam and as C++ libraries at https://github.com/JdMDE/jmatlib and https://github.com/JdMDE/ppamlib The package is useful for single cell RNA-seq studies but it is also applicable in other contexts where clustering of large data sets is required.

Project description:The objectives of this study were (i) to describe staphylococcal isolates recovered from bulk-tank raw milk collected from sheep and goat farms during a countrywide study performed in Greece, (ii) to study management factors potentially associated with their presence in bulk-tank milk and (iii) to provide evidence regarding their association with the quality of the milk. In total, 312 staphylococcal isolates, recovered from samples of bulk-tank raw milk from 444 small ruminant farms in Greece, were evaluated in this work. The in vitro formation of biofilm by the isolates was tested by combining the findings of (a) culture appearance on Congo Red agar plates and (b) results of a microplate adhesion test. The most frequently identified species was Staphylococcus aureus (75 isolates); other frequently recovered species were S. simulans (44 isolates), S. equorum (34 isolates) and S. haemolyticus (26 isolates); in total, 23 species were identified. In total, 224 (71.8%) isolates were biofilm-forming and were recovered from the bulk-tank milk samples of 148 sheep flocks (45.5%) and 55 goat herds (46.2%). There was evidence of seasonality in the isolation of staphylococci: during spring, mostly biofilm-forming isolates were recovered, whilst during summer, mostly non-biofilm-forming isolates were recovered. Among farms applying machine-milking, the proportion of farms from which biofilm-forming isolates were recovered was higher where water with temperature < 50 °C or ≥90 °C was used to clean the milking parlour. In the multivariable analyses, for farms applying machine-milking, the temperature of the water emerged as the only significant variable (p = 0.024), whilst in farms applying hand-milking, the only tendency that emerged was for the frequency of collection of milk from the farm tank (p = 0.08). In sheep flocks, recovery of biofilm-forming staphylococci from the bulk-tank milk was associated with higher somatic cell counts and higher total bacterial counts in the milk. The study identified abiotic factors related to the presence and isolation of these bacteria, specifically the temperature of water used for the cleaning of the milking parlour (in farms where machine-milking is applied) and the frequency of milk collection from the farm tank. These factors apply after the production of milk, and they could thus be regulated appropriately in order to reduce bacterial load and improve the quality of milk delivered to dairy plants. In sheep farms, an association was also seen between recovery of biofilm-forming staphylococci and high somatic cell counts in milk.

Project description:Staphylococcus aureus is a common causative agent of bovine mastitis in dairy herds. The emergence of methicillin-resistant Staphylococcus aureus (MRSA) in hospitals as well as the community is a significant and costly public health concern. S. aureus-related bovine mastitis is a common reason for therapeutic and/or prophylactic use of antibiotics on dairy farms. In this study, herd prevalence of S. aureus, including MRSA, was estimated from bulk tank milk (BTM) from Minnesota farms. A total of 150 pooled BTM samples from 50 farms, collected over 3 seasons (spring, summer, and fall of 2009), were assessed. Herd prevalence of methicillin-susceptible S. aureus (MSSA) was 84%, while MRSA herd prevalence was 4%. A total of 93 MSSA isolates and 2 MRSA isolates were recovered from 150 BTM samples. Antibiotic susceptibility testing of S. aureus isolates showed pansusceptibility in 54 isolates, resistance to a single antibiotic class in 21 isolates, resistance to two antibiotic classes in 13 isolates, and resistance to ?3 antibiotics classes and thus multidrug resistance in 5 isolates. The two MRSA isolates displayed resistance to ?-lactams, cephalosporins, and lincosamides and were multiresistant. Staphylococcal protein A gene (spa) typing identified spa types t529 and t034 most frequently among methicillin-susceptible isolates, while t121 was observed in MRSA isolates. Seven isolates, including the two MRSA isolates, produced staphylococcal enterotoxins B, C, D, and E on overnight culture. MRSA isolates were further genotyped using multilocus sequence typing (MLST) and pulsed-field gel electrophoresis (PFGE). Of the 2 MRSA isolates, one had a composite genotype profile of MLST ST 5-PFGE USA100-unknown spa type, which has been reported among hospital-associated MRSA isolates, while the second isolate carried the MLST ST 8-PFGE USA300-spa type t121 genotype, commonly identified among community-associated MRSA isolates. These results suggest that MRSA genotypes associated with hospitals and community can be isolated from milk at very low rates.

Project description:A total of 163 S. aureus isolates; 113 from mastitic milk (MM) and 50 from bulk tank milk (BTM) (2008, 2013-2015) submitted for bacteriologic analysis at the Penn State Animal Diagnostic Laboratory were examined for their phenotypic and genotypic characteristics. Multi-locus sequence typing (MLST) analysis identified 16 unique sequence types (STs) which belonged to eight clonal complexes (CCs). Majority of the isolates were variants of CC97 (68.7%) and CC151 (25.1%). CC97 comprised of seven STs, of which two were new STs (ST3273, ST3274), while CC151 comprised of three STs of which ST3272 was identified for the first time. Several farms had more than one ST type that were either members of the same clonal complex or unrelated STs. On one farm, six different STs of both categories were seen over the years within the farm. It was observed that ST352 and ST151 were the two main clonal populations in cattle not only in Pennsylvania but also globally. Most isolates were susceptible to all the antibiotics evaluated. 6.7% of isolates showed resistance to vancomycin and penicillin. Two isolates of ST398 showed multidrug resistance (>3 antibiotics) against clindamycin, erythromycin, tetracycline, and penicillin. It was noted that 59 of 163 (36.2%) isolates encoded for enterotoxigenic genes. Enterotoxin genes seg/sei accounted for ~85% of enterotoxin positive isolates. Toxic shock syndrome gene tsst-1 alone was positive in two isolates (ST352, ST 2187). 97.5% of CC151 isolates were enterotoxin seg/sei positive. Most isolates were positive for lukED (95%) and lukAB (96.3%) leukotoxin genes. Bovine specific bi-component leucocidin lukMF' was present in 54% of isolates. A prominent observation of this study was the explicit association of lukMF' with lineages ST151 and ST352. In conclusion, the findings of the study, suggest that small number of S. aureus STs types (ST352, ST2187, ST3028, and ST151) are associated with majority of cases of bovine mastitis in Pennsylvania dairy farms. It was observed that one ST of S. aureus predominated in the herd and this ST can coexist with several other ST types of S. aureus strains. When STs were interpreted along with virulence, leucocidin genes and antimicrobial resistance, ST-variants allowed better interpretation of the S. aureus molecular epidemiologic findings specifically for tracing recurrence or persistence of infections in cow over time, among cows in the herd, and between herds in Pennsylvania.

Dataset Information

A machine learning approach using partitioning around medoids clustering and random forest classification to model groups of farms in regard to production parameters and bulk tank milk antibody status of two major internal parasites in dairy cows

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets