The Use of MSTAT-C in a Cotton Breeding Program

D. S. Calhoun


 
ABSTRACT

Electronic record keeping in a plant breeding program is generally much faster and less error-prone than manual record keeping. MSTATC, developed and distributed by Michigan State University, offers several options for breeding program record keeping, as well as experimental design and analysis tools. The two primary routines for record keeping are "BRseries" and "ACseries". ACseries uses: 1) a "Master Accession" file for permanent storage of germplasm information such as name, pedigree, and user-determined, numeric descriptors; 2) "Fn" files for temporary (updated every selection cycle) storage of information on segregating populations such as cross number, pedigree, generation, desciptors and partial selection history; and 3) "Trial" files for design, data recording and data analysis of replicated experiments (with lattice or randomized complete designs). Fn and Trial files are built by ACseries using information from the Master Accession file. BRseries is a more basic system that performs many of the same functions as ACseries, but does not integrate directly with routines for replicated experiments. MSTATC breeding functions require minimal user input in terms of programming and setting up required files and output format. However, file format, designation of cross numbers and selection history, and format of output such as field books and labels is fairly inflexible.

Three routines for automated experimental design are available. "EXPseries" generates randomized complete block designs and associated computer files for experiments with one or several treatment variables in factorial or split-plot arrangements. Treatment "levels", including cultivars, must be numeric and consecutive. "ACseries" and "VARseries" generate randomizations and computer files for single-factor (i.e. cultivar) tests with randomized complete block or lattice designs, using cultivar information included in permanent source files. All three routines can be used to generate field books and print labels for seed packets, plot identification, etc. Data entry for all three routines is similar and can be done manually or through transfer of ASCII files from other programs or data loggers. Data analysis is fairly user-friendly and self-explanatory.

For more information, write to : MSTAT-C/Michigan State University, A87 Plant and Soil Science Building, E. Lansing MI 48824.



Reprinted from Proceedings of the 1994 Beltwide Cotton Conferences pg. 705
©National Cotton Council, Memphis TN

[Main TOC] | [TOC] | [TOC by Section] | [Search] | [Help]
Previous Page [Previous] [Next] Next Page
 
Document last modified Sunday, Dec 6 1998