Programs maintained on bcf by Steve Hardies

Clustal W

Clustal W is the most commonly used method for multiple alignment of divergent sequences. It uses progressive alignment, which means that closer sequences are aligned first, and prealigned sequences are represented by profiles during alignment. The implementation has a variety of alternatives for improving gap placement guided by secondary structure or arbitrary user intervention. Clustal is found in some form in many other packages (eg. GCG pileup), or at various web servers, although often restricted in some way. The interface is menu driven and each screen has context-specific help.

• Status:
• The currently installed version is 2.05
• The number of sequences handled by Clustal W depends on the amount of memory available in the executing computer. The bcf implementation exceeds tolerances somewhere above 250x700 characters.
• The SeaView alignment editor.  SeaView is currently out of service.
• The clustal implementation within the site-licensed LaserGene package is advertised as being rewritten to accept arbitrarily large alignments. We haven't benchmarked it yet, but that would normally be expected to slow it down.
• The current clustal W implementation includes the option to conduct NJ tree or UPGMA tree analysis with bootstrap support.  The documentation suggests that NJ is slow, but it should not be slow on bcf.  NJ is a much more accurate algorithm than UPGMA.
• A local help file is available with links to further documentation.

fastaq2phd

• This program processes a fasta file and its associated .qual file to a phd file
• It was was written by James D. White, University of Oklahoma, and retrieved from http://www.genome.ou.edu/informatics.html.
• We use it to represent a 454 contig as a single long phd file for reassembly with locally generated sequence reads.
• To access, export /home/hardies/genome/bin to your path.
• For options, type fastaq2phd
• Typical commands for the 454 contig conversion are:
• rename 454AllContigs.fna 454AllContigs; mkdir phd_dir; fastaq2phd -d phd_dir -p 454 454AllContigs
• To avoid 454 contigs becoming singlets, take a small section from ends and run fasta2phd.
• There is a script to harvest contig ends from a collection of contigs at the ou site.
• Technical notes concerning 454 data incorpration.

Phred/Phrap/Consed

Phred/Phrap and Consed are the Washington University Sequencing Center's shotgun sequence assembly and contig
      display programs. They feature probabilistic interpretation of chromatograms for quality of data, automatic contig
      assembly with heuristic upweighting of high quality data, and automated recommendataion of primers for finishing. The
      philosophy of these programs is to remove human inspection from the sequencing process as much as possible and to
      allow residual human inspection to zero in on problem areas as efficiently as possible. The system has a substantial
      learning curve. But in testing versus more user-friendly options, we find that this system ultimately saves significant time
      and expense.

Status:

• 7/2/2006 - Implemented version that can handle >64K phd files.
• Put /home/hardies/apps/genome in $PATH
• run phredPhrap_longreads instead of phredPhrap
• run consed_longreads instead of consed
• The purpose was to allow representing the 454 contract contigs as a single phd file containing consensus sequence and consensus quality values.
• This strategy may be used as a general way to pass forward the product of a previous assembly.
• See also fastaq2phd.
• 6/29/2006 - Implemented a separate version maintained by Steve Hardies
• Put /home/hardies/apps/genome in $PATH
• Default phredpar.dat file is in genome/lib
• Default vector.seq file is in genome/lib/screenlibs/
• determineReadTypes.perl is in genome/bin and implements Hardies' lab naming conventions.
• phredPhrap does not automatcally run determineReadTypes.perl. The user should access the phd_dir and run determineReadTypes.perl prior to running phredPhrap.  It is implemented this way to allow each user to run their own customized version of determineReadTypes.perl
• Installed version is Consed15 and phred version 0.020425.c
• Phred now handles CEQ data.
• consed is a pseudosym for consed_amd64_dyn.
• phrepar.dat has an entry for ABI 310.
• This implementation is used extensively by the Hardies lab, but mainly only phredPhrap and Consed.  Phred, addReads2Contig, and view_assembly have been briefly tested. Other functions may require additional configuration.
• 7/4/2006
• General system copy in /share/apps/genome, has been withdrawn from service pending reconciliation of its organization with the new system backup strategy.  It has become nonoperable.
• For the view assembly button, on small projects it will usually complain that all contigs have been excluded by user-selectable criteria. Dismiss that window, and look for another behind the main consed window, and decrease the number of reads and length of sequence required for view assembly to include the contig.
• Polyphred and autofinish have never been used at this site, and are probably not adequately configured.
• A local help file has been prepared.
• Phred/Phrap/Consed Home Page.
• Purpose of the vector.seq file (documentation).
• Technical notes concerning the maintenance history.