FASTLINK Home Page

What is FASTLINK?

FASTLINK is a software package to do the computations for genetic linkage analysis. Genetic linkage analysis is a statistical technique used to map genes and find the approximate location of disease genes. Since May 1993, we have been distributing faster versions of the existing genetic linkage analysis programs in LINKAGE 5.1. Several users have dubbed the new code "FASTLINK".

FASTLINK offers the following major features over LINKAGE. FASTLINK is roughly one order of magnitude faster than LINKAGE on long runs. FASTLINK permits the user to recover gracefully from a crash of the underlying computer. The FASTLINK distribution provides over 50 pages of new documentation beyond those that come with LINKAGE.

FASTLINK runs on all versions of UNIX, VMS, and DOS. For UNIX and VMS we distribute source code, and for DOS we distribute executable code.

Availability

We are currently distributing FASTLINK version 3.0P freely via anonymous ftp, available from Rice University's CS departmental FTP site (softlib.cs.rice.edu). After ftp'ing do cd pub/fastlink (for non-DOS versions) or cd pub/fastlink/dos (for DOS versions) There are two distribution versions:
a portable source version in fastlink.tar.Z
This version may be compiled for UNIX or VMS. It is distibuted as a compressed tar file. To unpack the archive on a UNIX machine, issue the commands 
     uncompress fastlink.tar.Z
     tar -xvf fastlink.tar

VMS users who do not have access to uncompress and tar equivalents can get the entire distribution piecemeal by ftp'ing to the server, logging in as user anonymous, and going to the directory pub/fastlink. Start with the file README, which will give you a roadmap to all the documentation.

DOS Versions with executable files
For DOS users, executable files are available uncompressed in the directory pub/fastlink/dos. Start with the file README.DOS for an explanation of which file is which.

Citations

Our initial improvements are described in the paper: 
  R. W. Cottingham Jr., R. M. Idury, A. A. Schäffer. Faster Sequential
  Genetic Linkage Computations, American Journal of Human Genetics 53(1993),
  pp. 252-263.
Most of the changes from version 1.1 to 2.0 are described in the paper: 
  A. A. Schäffer, S. K. Gupta, K. Shriram, and R. W. Cottingham Jr.
  Avoiding Recomputation in Linkage Analysis, Human Heredity, 44(1994),
  pp. 225-237.

We ask you to cite both papers if you use FASTLINK in any published experiments. Both papers are included in the distribution. FASTLINK is a modified version of LINKAGE. You should continue to cite the original papers on LINKAGE, if you used FASTLINK in a published experiment.

I am maintaining a mailing list of FASTLINK users. If you have retrieved the code and would like to be on the mailing list, send me e-mail at schaffer@cs.rice.edu.

Version 3.0P can also run in parallel. An early description of how we parallelize the ILINK program in FASTLINK can be found in: 

  S. Dwarkadas, A. A. Schäffer, R. W. Cottingham Jr., A. L. Cox, P. Keleher,
  and W. Zwaenepoel, Parallelization of General Linkage Analysis Problems.
  Human Heredity 44(1994), pp. 127-141.

Chris Hyams and I have subsequently adapted the parallel code to the best sequential version of FASTLINK and adapted the parallel algorithms to the LINKMAP and MLINK programs. Parallel FASTLINK runs either on shared-memory multiprocessors or on homogeneous networks of uniprocessors. Amazingly, we use essentially the same code for both types of hardware. This feat is made possible by using the TreadMarks distributed shared memory system, developed at Rice University. See the TreadMarks home page for more information.

Documentation

The complete set of FASTLINK documentation is now available on the web from the Genetic Linkage Analysis Web Server at Columbia University and the New York State Psychiatric Institute. The documentation can be found at http://linkage.cpmc.columbia.edu/soft/fastlink.html.

Future Development

Current efforts to improve FASTLINK include:
schaffer@cs.rice.edu
Last Update: 2/10/96