|Genetic Analysis Software|
Genetic linkage analysis is a statistical technique used to map genes and find the approximate location of disease genes. There was a standard software package for genetic linkage called LINKAGE. FASTLINK is a significantly modified and improved version of the main programs of LINKAGE that runs much faster sequentially, can run in parallel, allows the user to recover gracefully from a computer crash, and provides abundant new documentation. FASTLINK has been used in over 1000 published genetic linkage studies.
Version 1.0 of FASTLINK was first distributed in May 1993. We are now up to version 4.1P, first released in July 1999. You can download the current version with all documentation as well as some individual documentation files by following the hyperlinks on the left-hand side of this page. When you retrieve FASTLINK as a whole, you will get 1 file called fastlink.tar.Z. To unpack the archive on a UNIX machine, issue the commands
uncompress fastlink.tar.Z tar -xvf fastlink.tarUsers who do not have access to uncompress and tar equivalents can get the entire distribution piecemeal by ftp'ing to the server ftp.ncbi.nih.gov, logging in as user anonymous, and going to the directory pub/fastlink. Start with the file README, which is available by a direct hyperlink on the left to get a roadmap to all the documentation. If you want to see whether a new version of FASTLINK contains improvements that you want, look at the file README.updates, also available by a single click on the left. Thanks to David Stockton executables of the FASTLINK programs suitable for Macintosh OS X are available either by clinking on a hyperlink on this page or from the ftp site ftp.ncbi.nih.gov under directory pub/fastlink/mac. Other files of interest to Macintosh users are in that ftp directory.
The FASTLINK project is directed by Alejandro Schäffer. Over the years, participants in developing FASTLINK and providing customer service have included: Richa Agarwala (NIH) Ann Becker (Technion), Robert Cottingham Jr., Alan Cox (Rice U.), Sandhya Dwarkadas (now at U. Rochester), Dan Geiger (Technion), Sandeep Gupta (now at Trilogy), Christopher Hoelscher, Chris Hyams (now at Trilogy), Ramana Idury (now at Kiva Genetics), Peter Keleher (now at U. Maryland), K. Shriram (Rice U.) Willy Zwaenepoel (Rice U.). I am also very grateful to hundreds of users around the world for comments, questions, suggestions, and bug reports, and most of all for being bold enough to try some new software and stick with it.
The FASTLINK project has benefited from two interesting concepts from computer systems. First, K. Shriram implemented a checkpointing facility that allows the FASTLINK main programs to recover gracefully from a crash of the underlying computer. Stephen Rich (now at Wake Forest U.) gets the prize for the most unusual usage of checkpoint/crash recovery. Just before moving his workstation from Minnesota to North Carolina, he killed an ongoing FASTLINK run, and shut down the computer. Once he and the workstation got installed in North Carolina, he resumed the run near where it left off and completed it successfully. I have reused some of K. Shriram's checkpointing code in the PSI-BLAST module of the widely used BLAST package.
Starting with version 2.3P, the ilink, mlink, and linkmap programs in FASTLINK can run in parallel. What is especially surprising is that essentially the same code can run on either shared-memory multiprocessors or on homogeneous networks of uniprocessors. This feat is made possible by using the TreadMarks distributed shared memory system developed at Rice University.
FASTLINK comes with substantial documentation. Most of the documentation files are in ASCII text. Two of the more popular documents that are in PostScript are: The Mystery of (the) Unknown, which explains the preprocessor program unknown, and Loops in FASTLINK . You can download PostScript versions by clicking on the hyperlinks to the left or read the documents one page at a time by clicking on the hyperlinks in this paragraph. Thanks to many users and especially the student participants on the FASTLINK projects who kept asking questions that deserved clear and written answers for everyone's benefit.
To promote better understanding of how to do simple linkage analyses, I teach a class called "Cookbook Linkage Analysis, Ab Initio" . Follow the hyperlink in this paragraph to get to a page that lets you link to the course overheads and 11 recipes for different aspects of linkage analysis. Thanks to Jim Tomlin for putting the course materials on the Web.
I maintain a mailing list of FASTLINK users, used mostly to announce new versions, updates, and bug fixes. If you have retrieved the code and would like to be on the mailing list, send me e-mail at firstname.lastname@example.org.