What's New in Version 10.3

[Genhelp | Program Manual | User's Guide | Data Files | Databases | Release Notes ]

Table of Contents

Programs
New Programs
Programs Not Available For This Release

Enhancements
Program Enhancements

Bug Fixes
Program Bug Fixes
SeqLab Bug Fixes
Package-Wide Bug Fixes (LINUX Only)

New Programs

[ Top | Next ]

The programs listed below are new to version 10.3 of Accelrys GCG (GCG).

Database Searching

PSIBLAST

PSIBLAST, or Position-Specific Iterated BLAST, uses the methods described in Altschul, et al. Nucleic Acids Res. 25(17): 3389-3402 (1997) and Schaffer, et al. Nucleic Acids Res. 29(14): 2994-3005 (2001) to search for similarities between protein query sequences and all the sequences in one or more protein databases.

PSIBLAST uses position-specific scoring matrices (PSSMs) to score matches between query and database sequences, in contrast to BLAST, which uses pre-defined scoring matrices such as BLOSUM62. PSIBLAST may be more sensitive than BLAST, meaning that it may find distantly related sequences not found with a BLAST search.

Protein Analysis

TransMem

TransMem builds on the method of Sonnhammer et al. (Proc. of Sixth Int. Conf. on Intelligent Systems for Molecular Biology, 175-182 (1998)) to predict likely transmembrane helices in one or more input proteins. The method is based upon a Hidden Markov Model (HMM) that has been trained on a set of membrane proteins with helical membrane spanning regions.

Programs Not Available For This Release

[ Previous | Top | Next ]

The programs listed below are not available for version 10.3 of GCG on the Linux platform.

Evolution

PAUPDisplay

Because the version of the PAUP programs in GCG 10.3 is not compatible with Linux, all programs that call the PAUP package within GCG have been disabled on the Linux platform. We hope to be able to provide these programs on all platforms with the next release of the package.

PAUPSearch

Because the version of the PAUP programs in GCG 10.3 is not compatible with Linux, all programs that call the PAUP package within GCG have been disabled on the Linux platform. We hope to be able to provide these programs on all platforms with the next release of the package.

Program Enhancements

[ Previous | Top | Next ]

Database Searching

NetFetch

The mechanism used by NetFetch to connect to NCBI's Entrez server has been modified such that the parameter -URL has no effect. In addition, all input sequences must now be of the same type (i.e., either nucleotide or protein).

Database Utilities

          GCGToBLAST

 

GCGToBLAST now uses the new parameter -PARSEseqid by default. Databases formatted with this parameter will consist of two additional index files, which permit BLAST and PSIBLAST to correctly display sequence names in any of the multiple alignment output formats. This behavior can be suppressed by specifying -NOPARSEseqid.

Gene Finding and Pattern Recognition

FindPatterns

Enhancement: A new command line parameter, -LIStfile, can now be used to produce list file output, in addition to the normal output file. This parameter replaces -NAMes, which will be removed in a future release. To ensure compatibility with earlier versions, -LIStfile has no effect if -NAMes is also specified.

Motifs

Enhancement: A new command line parameter, -LIStfile, can now be used to produce list file output, in addition to the normal output file. This parameter replaces -NAMes, which will be removed in a future release. To ensure compatibility with earlier versions, -LIStfile has no effect if -NAMes is also specified.

Program Bug Fixes

[ Previous | Top | Next ]

Database Searching

NetFetch

Workaround: When retrieving a long sequence using NetFetch, add the -RAW parameter to the command line. This will additionally produce a GenBank sequence file containing the sequence. Now you can run FromGenBank to produce several small single source files of your sequence.

Gene Finding and Pattern Recognition

CodonPreference

Problem: The codon preference statistics for each window determined by CodonPreference were calculated using one too many codon preference statistic values in the sum. That is, instead of using the formula
P = exp (sum over window ln (p)) / window
P was calculated using
P = exp (sum over window+1 ln (p)) / window.

As a result, if you asked for the window preference statistics with -PWINdow=3, the sum over the codon preference statistics of the 3 codons in a window plus the next codon were used instead.

Update: CodonPreference now calculates the correct codon preference statistics for each window.

FindPatterns

Problem: In RSF output, matches longer than 350 were always reported as features that were 350 positions in length.

Update: Matches are now correctly reported in RSF output.

Importing and Exporting

BreakUp

Pairwise Comparison

FrameAlign

Problem: If you ran FrameAlign on the Irix or Solaris platform with the parameters -GLObal and -ENDWeight set, the results would contain incorrect alignments and quality scores.

Update: You can now use both -GLObal and -ENDWeight and the program will produce the correct results on all supported platforms.

Primer Selection

Prime

Problem: If you ran Prime and selected a range of the input sequence to be analyzed, the resulting RSF file would contain only the selected range of the input sequence, but the feature positions correlated to the entire input sequence. Therefore, the features listed in the RSF file would not match onto the sequence in the RSF file.

Update: The RSF file now contains the entire input sequence and the feature positions are correct.

Utilities

PrositeToGCG

Problem: If you ran PrositeToGCG on the Solaris platform, the output files would be incomplete and have bad file names.

Update: PrositeToGCG now produces a complete set of output files with valid names.

Translation

Translate

Workaround: Suppress interactive mode by including the -Default parameter on the command.

SeqLab Bug Fixes

[ Previous | Top | Next ]

Problem: Deletion of tildes from the 5' end of a sequence in the SeqLab editor caused the edited sequence in the trace viewer to shift relative to the trace and the raw sequence. Reinserting tildes did not shift the edited sequence back into register.

Update: SeqLab now maintains proper registration between the original base calls and the trace peaks in the presence of alignment offsets.

Editor Mode (LINUX Only)

Workaround: Use the Control key in combination with the right mouse button, or use the Select by Name or Select Range buttons under the Edit menu to select individual entries or several non-contiguous sequence residue stretches.

Workaround: After switching to a different mode, click the Protect button to check that the protections are set correctly, then click the OK button. The selected editing mode should work.

Programs Run Through SeqLab

MEME

Problem: If you selected the "Exactly one occurrence of each motif in each sequence" option when analyzing a protein sequence, the -TWOStrands parameter was automatically added to the command line. Then, when trying to run the program, it would fail with the log message "You must use default DNA alphabet if using complementary strands!"

Update: Now when you select "Exactly one occurrence of each motif in each sequence" with a protein sequence, MEME runs correctly, without adding the -TWOStrands parameter.

ProfileMake

Problem: If you ran ProfileMake and selected the "linear weighting" option, the program would run using the default "exponential weighting" option instead.

Update: Now the "linear weighting" option works correctly.

StemLoop

Problem: The main window for StemLoop would select "see the stems" by default, and would also provide an option to "see the stem coordinates." Both of these options would display the output of the StemLoop run in the Job Manager window.

Update: The StemLoop window now contains the options "file the stems" and "file the stems as points for DotPlot," which produce files containing the results in the Output Manager.

Package-Wide Bug Fixes (LINUX Only)

[ Previous | Top ]

Known Bug: If you run one of the multi-threaded programs that report CPU times in the summary (such as the FastA family and FrameSearch), those times will be incorrect.

Workaround: You can still determine the correct process time by using the "time" system switch. For example, use the following command line to start FrameSearch and get the correct process time:

$time framesearch

After entering this command, a line displays before the next system prompt. The first value in this line provides the process time in seconds.

[Genhelp | Program Manual | User's Guide | Data Files | Databases | Release Notes ]

Technical Support: support-us@accelrys.com, support-japan@accelrys.com,
or support-eu@accelrys.com

Copyright (c) 1982-2005 Accelrys Inc. All rights reserved.

Licenses and Trademarks: Discovery Studio ®, SeqLab ®, SeqWeb ®, SeqMerge ®, GCG ® and, the GCG logo are registered trademarks of Accelrys Inc.

All other product names mentioned in this documentation may be trademarks, and if so, are trademarks or registered trademarks of their respective holders and are used in this documentation for identification purposes only.

www.accelrys.com/bio