Transitioning from GCG to EMBOSS
on the Helix Systems

----------------------------------------------------------------------
                     Welcome to the WISCONSIN PACKAGE 
                           Version 10.3-UNIX
                             Installed on irix64
 
                 Copyright (c) 1982 - 2001, Accelrys Inc.
 A wholly owned subsidiary of Pharmacopeia, Inc. All rights reserved.

          Published research assisted by this software should cite:
       Wisconsin Package Version 10.2, Accelrys Inc., San Diego, CA
----------------------------------------------------------------------

GCG program groups

• Comparison
• Database Searching
• Editing and Publication
• Evolution
• Fragment Assembly
• Gene Finding and Pattern Recognition
• Importing and Exporting
• Mapping
• Primer Selection
• Protein Analysis
• RNA Secondary Structure
• Translation
• Utilities

European Molecular Biology Open Software Suite

EMBOSS is a free Open Source software analysis package specially developed for the needs of the molecular biology (e.g.EMBnet) user community.

EMBOSS program groups

• Alignment (Consensus, Differences, Plots, Global, Local, Multiple)
• Display
• Edit
• Enzyme Kinetics
• Feature Tables
• Information
• Nucleic (2D structure, Codon usage, Composition, CPG islands, Gene finding, Motifs, Mutation, Primers, Profiles, Repeats, Restriction, Transcription, Translation)
• Phylogeny
• Protein (3D structure, Composition, Motifs, Mutation, Profiles)
• Utils

• Large collection of programs that cover most areas of sequence analysis.
• Most GCG programs have an EMBOSS equivalent and vice versa. (but not all!)
• Documentation & sample inputs/outputs available.
• Web or command-line interfaces
• Same databases available.
• Command-line is good for large-scale batch processing.
• Can create your own databases.

• EMBOSS accepts all sequence formats!!!
• EMBOSS accepts multiple sequences in an input file.
• GCG is commercial, EMBOSS is open-source.
• EMBOSS allows you to develop/add your own programs.
• GCG has better support (?)
• GCG has more database divisions.
• GCG has better sequence search capabilities.
• EMBOSS is more frequently updated.

Most GCG programs have an EMBOSS equivalent and vice versa. (but not all!)

http://helix.nih.gov/Applications/emboss-gcg.html

GCG	EMBOSS
Helix Systems, CIT NHGRI NICHD Codon.nih.gov at NIMH NCI, Frederick	Helix NCI, Frederick

Cost.
Administrative workload.
Decreasing usage.
Equivalent package in EMBOSS.
EMBOSS is open-source.
No license issues with EMBOSS.

How to access	GCG	EMBOSS
Web	GCG-Lite http://helixweb.nih.gov/emboss-lite No username or password required Open to all NIH machines Now renamed to EMBOSS-Lite.	Will transition to using EMBOSS programs behind the scenes. No username or password required Open to all NIH machines
Web	SeqWeb (shut down in Dec 05) Required Helix account and login/password Open to all NIH machines.	EMBOSS full-function geb interface at http://helixweb.nih.gov/emboss No username or password required Open to all NIH machines
Command-line	on Helix & Nimbus (Helix account required)	on Helix, Nimbus, Doublehelix and the Biowulf cluster. (Helix account required)

To run EMBOSS on the Helix/Nimbus/Doublehelix command-line:

1. type 'emboss' to initialize the package
2. then type the name of an EMBOSS program

(demo)

GCG has a powerful sequence locator, 'lookup'. EMBOSS does not. Only the accession number and the ID lines are indexed in the EMBOSS databases.

If you know the accession number of your sequence, you can find it in EMBOSS.

If you don't know the accession number, search on the NCBI website (http://www.ncbi.nlm.nih.gov)

Demo: searching for 'fibronectin' sequences.
GCG-Lite: http://molbio.info.nih.gov/molbio/gcglite.

1. Choose 'Sequence Lookup - simple query'.
2. Enter 'fibronectin' in the search box.
3. Click on the accession number for one result. The result page will show the sequence in GCG format.
4. Click on the browser menu File -> Save As, and save the sequence onto your desktop. Important: save it in Text Only Format.

Finding sequences via NCBI Entrez: http://www.ncbi.nlm.nih.gov/

1. In the top box, select a database to search (Protein or Nucleotide)
2. Enter fibronectin in the box and click Go
3. Click on one of the hits to display the sequence in Genbank format.
4. Click on Send to -> File, and save the sequence to a file in the default Genbank format. Rename it!!!
5. Or Select 'Fasta format' in the Display menu on the NCBI web page. Click on Send to -> File and save the sequence in a file in Fasta format. (Rename it!!)

GCG accepts
-GCG
- MSF
- RSF

EMBOSS accepts
- EMBL
- Genbank
- SwissProt
- PIR
- GCG
- MSF
- Clustal
- Plain, raw
- and many more.

GCG: one sequence per GCG file
Input: only first sequence in a file is read.
Output: into multiple files, one per sequence

EMBOSS: one or more sequences per file
Input: all sequences are read and processed
Output: all sequences into one file, unless you use -ossingle.

Demo: Running an EMBOSS program with a GCG-format sequence.
GCG: fetch Genbank:Ax700501
GCG: map ax700501.gb_pat
EMBOSS: remap ax700501.gb_pat

All GCG-format sequences should work transparently in EMBOSS!

             GCG Databases available: 
                   GenBank             Release 156.0 (10/06)
                   GenPept             Release 156.0 (10/06)
                   PROSITE             Release  20.3 (01/07)
                   Restriction Enzymes (REBASE)  701 (01/07)
                   Pfam                Release    21 (11/06)
                   SWISS-PROT          Release 51.4 + updates to 09/Jan/07
                   GP_New              FROZEN on 23/Dec/06 (117,658 entries)
                   GB_New              FROZEN on 21/Dec/06 (3,210,047 entries)

  EMBOSS Databases available: 
  genbank             Release 157 (18/Dec/06)
  genpept             Release 157 (26/Dec/06)
  est                 Release 157 (18/Dec/06)
  refseqaa            Release 21 (11/Jan/07)
  refseqnt            Release 21 (11/Jan/07)
  PROSITE             Release 20.3 (09/Jan/07)
  Restriction Enzymes (REBASE) 701 (30/Dec/07)
  Transfac            Release 10.4 (15/Dec/06)
  prints              Release 38_0 (21/Sep/05)
  uniprot             Release 9.4 (09/Jan/07)
  gpnew               21/Jan/07, 215603 entries since 26/Dec/06 rel 157
  gbnew               21/Jan/07, 1585034 entries since 18/Dec/06 rel 157

How to identify sequences in a database:
GCG: db:accession (e.g. GB_VI:HIM010487 or SWISSPROT:CCNT2_HUMAN)
EMBOSS: db:accesssion (e.g. genbank:AJ010487 or uniprot:CCNT2_HUMAN)
Note that the database names are different, and only accession numbers are indexed in EMBOSS.

How to use a database sequence in a program:
GCG: translate GB_VI:HIM010487
EMBOSS: transeq genbank:AJ010487

Demo:
GCG: fetch swissprot:cram_craab
EMBOSS: seqret uniprot:cram_craab (in default Fasta format)
EMBOSS: seqret -outseq=gb::cram_craab.gb uniprot:cram_craab (in Genbank format)

GCG is able to access subsets of databases (e.g. gb_ba, gb_vi), or combinations of databases (GB+ = Genbank + EST). This is more complex in EMBOSS.

Some GCG programs can produce graphic output, such as a Postscript or PDF file or Xwindows graphic output. Direct printing to a Postscript printer is also possible. List of GCG programs that can produce graphics output.

Old GCG 'figure' files will not be readable by EMBOSS programs.

Equivalent EMBOSS programs can produce Postscript or png files, or Xwindows graphic output (other options also possible). There is no EMBOSS equivalent to the GCG 'setplot' command; the output type can just be specified on the command line.

Sample EMBOSS session for the 'pepwheel' program (the equivalent to GCG's 'helicalwheel').

% pepwheel
Shows protein sequences as helices
Input protein sequence: cram_craab.swissprot
Graph type [x11]: ps
Created pepwheel.ps
% ps2pdf pepwheel.ps pepwheel.pdf

Postscript file produced by EMBOSS pepwheel.
PDF file created by ps2pdf above.

GCG-Lite (http://molbio.info.nih.gov/molbio/gcglite) is a simple intuitive web interface to the most commonly used GCG programs.

Replaced by:
- GCG-Lite (name change!) running EMBOSS programs behind the scenes.
- EMBOSS full function web interface (http://helixweb.nih.gov/emboss)

Demo of GCG-Lite running EMBOSS-Lite.
Demo of same program in EMBOSS web interface.

Demo: gap of genbank:M25079 and genbank:M25113
Demo: needle of the same sequences.

To find EMBOSS programs for a particular task:

1. (web) See the list of equivalent programs in GCG & EMBOSS
2. (web) Try the EMBOSS help ( http://helixweb.nih.gov/emboss, then click on 'Help' and search for a text phrase)
3. (web) Browse the list of all EMBOSS programs (http://helixweb.nih.gov/emboss, then click on 'Program Docs')
4. (command-line) Use 'wossname' to find programs relating to a topic. (e.g. 'wossname restrict')

To find command-line options:

1. Keep a browser open to the EMBOSS help http://helixweb.nih.gov/emboss/html/
2. Use 'command -help -verbose'. (e.g. 'remap -help -verbose')

Demo: EMBOSS: find restriction enzyme cleavage sites.

1. How do you get a list of enzymes?
2. How do you find more about an enzyme?
3. How do you know what parameters are reasonable?
4. How do you know if your output is normal?
5. How do you interpret your output?

Answers:

1. listed at the end of a 'remap' run.
2. Use the 'redata' program.
3. Look at the help pages for that program. http://helixweb.nih.gov/emboss/html/, or click on 'Help'. e.g. search for 'restrict'.

More help:
Email the Helix staff at staff@helix.nih.gov
Call the Helix staff at 301-594-6248 (the NIH Helpdesk - ask for Helix staff regarding a GCG or EMBOSS question).

• Blast
• Fasta
• BLAT
• Wu-Blast
• MFOLD
• ClustalW
• HMMER? (available on Biowulf, email staff@helix.nih.gov if you need this).

Use GCG-Lite (EMBOSS-Lite) for occasional use.
Use the full EMBOSS website if you need additional programs.
Use EMBOSS on the Helix command line for large-scale analysis (e.g. ~50 sequences).
Use EMBOSS on Biowulf for very large-scale analysis (e.g. 100+ sequences).