About TFAM
TFAM is a BioPerl-based tRNA function classifier that uses sequence profile models to classify the function of unknown query tRNAs. It is useful for classifying initiator tRNAs, some bacterial lysidinylated isoleucine tRNAs, and the charging identity of suppressors, pseudo-tRNAs and other tRNAs with unusual structure.
Download and Licensing of TFAM
From this page you can obtain a GNU tarballed and gzipped or a zipped version of TFAM version 1.4.
Since TFAM 1.02 TFAM is released under the Gnu General Public License version 2.
TFAM Web-Server
This page provides access to the stand-alone version of TFAM for UNIX and UNIX-related systems. A TFAM Webserver is down for the moment and will be up again in early 2016.
Citing TFAM
For more information or if you should publish analyses with or modifications of TFAM, please see or cite:
D.H. Ardell and S.G.E. Andersson (2006). TFAM detects co-evolution of tRNA identity rules after lateral transfer of Histidyl-tRNA synthetase. Nucl. Acids Res. 34(3):893-904.
More Information, Versions and Compatibility
The current standard TFAM models are based on the year 2000 release of the Sprinzl tRNA database (the last release with a majority of experimentally annotated tRNAs). Users can build and define their own custom TFAM models.
NEW as of July, 2012 TFAM 1.4 is released. This release fixes a compilation bug with the COVEMF subpackage that arose by a naming conflict with stdio.h.
TFAM 1.3 added the ability to auto-detect anticodons in the input data and use them to detect anticodon-based conflicts with its own classifications.
In Feb. 1007, TFAM 1.0 was released as both of a Web Server and standalone interface with simplified installation. Earlier installation versions of TFAM depended on The Perl Data Language (PDL). This dependency is now removed (thanks to contributions from Helena Tåquist). TFAM 1.0 added an improved bacterial model (version 0.2) with the ability to detect lysylated isoleucine tRNAs in some phyla of bacteria (Thanks to work by Yuanyuan Cui and data generously provided in 2004 from Paul Higgs), and two preliminary eukaryotic and archaeal models (versions 0.1) specifically tested and released for the purpose of classifying initiator tRNAs. The use of eukaryotic and archaeal 0.1 TFAM models for any other class except initiator tRNAs has not been tested in any way!
TFAM has been used and cited in many genome projects and in the tRNAdb-CE database as well as in a 2006 study on classifying tRNAs in bacteria. The authors of that study published a third-party TFAM model that can be used with TFAM 1.0. For more information see Silva et al. (2006) Differential annotation of tRNA genes with anticodon CAT in bacterial genomes. Nucleic Acids Res., 34, 6015-6022.
Anybody can create tRNA classification models that can be used with standalone interface of TFAM. See the README for more information on this and using TFAM in general.
TFAM is known to work on UNIX type systems, but has been only tested on Linux and Mac OSX. Portability of TFAM to Windows systems is unfortunately unlikely, because it depends on the COVEMF package. COVEMF is shipped together with TFAM, you don’t need to download it separately.