Sequence evaluation of four pooled-tissue normalized bovine cDNA libraries and construction of a gene index for cattle.

An essential component of functional genomics studies is the sequence of DNA expressed in tissues of interest. To provide a resource of bovine-specific expressed sequence data and facilitate this powerful approach in cattle research, four normalized cDNA libraries were produced and arrayed for high-throughput sequencing. The libraries were made with RNA pooled from multiple tissues to increase efficiency of normalization and maximize the number of independent genes for which sequence data were obtained. Target tissues included those with highest likelihood to have impact on production parameters of animal health, growth, reproductive efficiency, and carcass merit. Success of normalization and inter- and intralibrary redundancy were assessed by collecting 6000-23,000 sequences from each of the libraries (68,520 total sequences deposited in GenBank). Sequence comparison and assembly of these sequences was performed in combination with 56,500 other bovine EST sequences present in the GenBank dbEST database to construct a cattle Gene Index (available from The Institute for Genomic Research at http://www.tigr.org/tdb/tgi.shtml). The 124,381 bovine ESTs present in GenBank at the time of the analysis form 16,740 assemblies that are listed and annotated on the Web site. Analysis of individual library sequence data indicates that the pooled-tissue approach was highly effective in preparing libraries for efficient deep sequencing.

A conceptual database model for genomic research.

We describe a conceptual model for genome databases that facilitates the process of building, maintaining, and disseminating physically anchored genetic linkage maps. The model has been implemented as a relational database at the Roman L. Hruska U.S. Meat Animal Research Center (MARC). Development of consensus maps using disparate data from different reference pedigrees or laboratories is supported. The model is of use to quantitative and population geneticists interested in loci that affect phenotypes and marker-assisted selection, and it is sufficiently flexible for centralized, species genome databases facilitating comparative mapping. The MARC genome database is used to assemble, maintain, and disseminate physically anchored genetic linkage maps for cattle, swine, and sheep currently based on more than 100,000 genotypes from 1,000 markers. Integrated with linkage analysis software, this database permits frequent updates of physically anchored genetic linkage maps.

Predicting the growth of Salmonella typhimurium on beef by using the temperature function integration technique.

Lag and generation times for the growth of Salmonella typhimurium on sterile lean beef were modeled as functions of cooling time under various carcass-chilling scenarios. Gompertz growth models were fit to the log10 colony counts over time at each of six temperatures in the range of 15 to 40 degrees C. Lag and generation times were defined as the points at which the second and first derivatives, respectively, of each growth curve attained a maximum. Generation time and lag time parameters were modeled as functions of temperature by use of exponential-decay models. The models were applied to typical beef carcass-cooling scenarios to predict the potential growth of S. typhimurium during the cooling of beef. Validation studies indicated no significant difference between the observed and predicted bacterial populations on inoculated lean and fatty beef tissues cooled at either 6 or 9 degrees C/h.

Principal gastrointestinal variables associated with metabolic heat production in pigs: statistical cluster analyses.

The purpose of this study was to classify organs/tissues into clusters according to their association with fasting heat production (FHP) by multivariate cluster analysis and to identify primary tissues. A data set consisting of 47 variables and 2256 data points was collected from 48 growing male pigs. Variance was perturbed by varying the daily feed allowance. Cluster analysis was conducted at each of six levels of body dissections, and a final analysis was conducted with all 47 variables. Small intestine, pancreas, liver, and kidneys were consistently assigned to the FHP-cluster; stomach, cecum, and colon to clusters correlated, r greater than 0.5, to the FHP-cluster; carcass, thoracic and offal variables to unrelated clusters. In the final analysis with all 47 variables represented, only variables from the small intestine and jejunum were assigned to the FHP-cluster. It is concluded that the small intestine, which accounts for only 2% of the body weight, is a primal organ system in swine and that pancreas, liver, and kidneys are of similar importance.

Production responses in early lactation to additions of soybean meal to diets containing predominantly corn silage.

Responses in yield of milk and intake of dry matter to increased concentrations of dietary crude protein by cows in early lactation were modeled from published data. The data base was 625 Holstein cows in second or later lactation fed diets primarily composed of corn grain and a combination of corn silage and alfalfa. Dietary crude protein as a percent of dry matter ranged from 9.5 to 20.2% and was varied with soybean meal. Exponential models were developed to predict changes in milk yield and dry matter intake per unit change in dietary crude protein. The consequence of changes in concentration of dietary crude protein on performance can be predicted with the models. The dose-response relationships can be used with relevant price information to estimate the concentration of crude protein that maximizes returns over feed costs. Profitability is improved by adjusting concentration of dietary crude protein for changes in the price of soybean meal. Present feeding standards could be enhanced by incorporation of information on dose-response and economics.