Plant & Animal Genome VI Conference

W112

JAVA, A PROGRAMMING SOLUTION TO MULTI-PLATFORM QTL ANALYSIS USING REGRESSION TECHNIQUES

1. School of Biological Sciences, University of Birmingham, Birmingham B15 2TT. UK.
2. Dept. of Cereals Research, The John Innes Centre, Colney Lane, Norwich NR4 7UH. UK.
3. Roslin Institute (Edinburgh), Roslin, Midlothian, EH2 9PS. UK.
4. Dept. of Computer Science, University of Warwick, Coventry CV4 7AL. UK.

The original objective of this project was to create a single software package integrating a comprehensive suite of tools for QTL analysis & visualisation in plant & animal populations. One prerequisite was that the software should run on multiple platforms, including at least IBM PC compatibles, UNIX and Macintosh, using a "windows" graphical user interface. In the first instance the QTL mapping methods of group members were to be included namely "flanking marker regression" (Haley & Knott, 1992) and "multiple marker regression" (Kearsey & Hyne, 1994), other methodologies to be included as time allows. These QTL analysis methods would be augmented with routines for file input, data manipulation, graphical representation of results, result/data export to other applications and a comprehensive help system for the novice user.

The project was initiated with a view to cross-platform development i.e. developing the software for the PC compatible (using C++) then port it to the other platforms. However, a short feasibility study identified the object-based programming language Java as suitable for the project. Java has the advantage of being a multi-platform language, i.e. it will run on all platforms for which a Java Virtual Machine is available (already including PC, UNIX and Macintosh). The developing applet software is available to run across the Internet for evaluation (http://web.bham.ac.uk/g.g.seaton) using browsers (Netscape Navigator & Microsoft Internet Explorer) and appletviewers. A comprehensive help/manual will be produced in HyperText Markup Language (HTML) and will be linked to the same WWW site. It is planned that the software will ultimately be made available through file transfer protocol (FTP) for copying to machines without direct Internet access.

REFERENCES

• Haley, C. S. & Knott, S. A. 1992. A simple regression method for mapping quantitative trait loci in line crosses using flanking markers. Heredity, 69, 315-324.
• Kearsey, M. J. & Hyne, V. 1994. QTL analysis: a simple "marker regression" approach. Theor. Appl. Genet., 89, 698-702.