Scaffold 2.2.1 release notes:

Scaffold 2.2.1 fixes an issue where indexing GOA database on 64 bit
windows and mac machines was very slow.  This is the only change in this
version and users on 32 bit machines need not upgrade.  (1714)

Scaffold: 2.2.0 release notes:

* Scaffold now supports loading protein annotations from the Gene Ontology Annotations (GOA) database. In addition to retrieving annotations from the NCBI, Scaffold can now annotate against a local database that is more extensive (particularly for the UniProt database) and includes information about the provenance of the annotation (for example, whether it was made by a human or a computer program). Scaffold also now allows you to choose exactly which GO terms are displayed.
(1593-1595 and others)

* Re-written "Queue files from Mascot Server" feature allows filtering and sorting, and improves performance and reliability. (1577)

* ScaffoldBatch 2.2 supports platform independent paths in scafml (batch
driver) files. In other words:

<Scaffold pathsep="/">
<!-- ... -->
<InputFile>../foo/bar/baz.dat</InputFile>
<!-- ... -->
</Scaffold>

should work on Windows. (1473)

* A bug where Scaffold could, in some rare cases, loose protein isoforms has been fixed. In order to restore these isoforms, Scaffold will automatically re-perform protein grouping on file open. (1691)

* Improved collation of spectra across multiple search engines, especially for non-mudpit searches. (1588)

* A bug where certain Mascot files could cause Scaffold to erroneously register very large numbers of modifications has been fixed. (1616)

* Scaffold now attempts to recover data files after a crash. (1673)

* Fixed a bug where Scaffold 2.x incorrectly detected ETD/ECD data in Mascot files. (1692)

* Performance improvements made to looking up proteins in large FASTA databases. (1693)

* Restored output of modified amino acids in spectrum report. (1615)

* Updated the set of gene ontology terms that Scaffold uses. (1634)

* Fixed a problem with Spectrum Mill modifications. (1644)

* Added "End of File" to all Scaffold exports. (1664)

* Converting version 1.x files is now faster. (1395)

Scaffold 2.1.2 release notes:

- Uncheck visible checkbox in Scaffold's Samples view now correctly unchecks one protein, not two.

- SRF file parent error tolerance is now properly read by Scaffold.

- Percent coverage display in Scaffold's Samples view now works.

- Reset peptide validation option has been replaced after being removed. It is located under the "Experiment" option on the task bar.

Version 2.1.1

-File merging now works.

-Thinning no longer lowers protein probabilities.

-Venn Diagram numbers now correctly represent the number of protein identifications.

-Correctly read error tolerance from *.srf files.

-Fixed bug in View drop down menu so that Show lower scoring matches filter works
-Canceling a Scaffold analysis during the collation step will no longer cause errors in the resulting Scaffold file.
-When loading mascot data from the server, we now save the data in the SFD file until it has been analyzed.
-Various fixes for Spectrum Mill (collation, load times, etc.

Version 2.1.0

Protein probability updated
Adding NCBI annotations now turns on Save option
Wait icon in Samples view now gives correction indication of busy system
In Edit > Preferences > Processors, Scaffold is now able to choose the number of
available processors.
Condense Data On Load capability is now operational
Collation now operational
Load from Directory with Unicode/MCBS now functional
Relative paths in scafml now functional
Updated Converter
ProtXML now exports correct calc_neutral_pep_mass for peptides

Version 2.0.5

Mascot file loading no longer stops if a single file has an error -- unless that is the only file we are loading.
Able to unlock scaffold via E-Mail.
Changed command line arguments to deal with garbage collection better.
Eliminated spurious Out of Memory errors.
Fixed C and N term modifications are applied only to the end termini and not throughout the peptide for SEQUEST files.
C and N termini are given as deltas for SEQUEST.
SEQUEST out files now have lines ignored they are not parse-able.
Improved error handling of broken files. We now skip only the files that we can not parse rather then halting on a bad file.
We now properly parse the decoy_peptide section of the Mascot file and add in the reverse hits under different protein names.
Properly identify variable modifications in Mascot data.
SRF Files now have their accession numbers parsed correctly.
Tar files for SEQUEST now have all their content read.
Memory is now able to be set at 4x the physical memory in cases where Java doesn't see all the memory.
Load cache objects sooner, this increases performance and helps with batch processing.

Version 2.0.3

Fixed bug where SEQUEST files in tar.gz format would not be read properly.
Fixed bug in Scaffold reader where some scientific notation could cause the parser to crash.
Fixed bug in Mascot reader where new contentTypes could in certain cases confuse the parser.
Fixed bug with multiple Amino Acid issue in SEQUEST parser.
Fixed bug where converted files would lose the name of their database.
Fixed bug where database parsing was broken. Could not load valid databases, only invalid databases.

Version 2.0.2

General: added new 64 bit linux JRE for proper 64 bit execution
General: fixed execution bug when running batch mode
Fixed null pointer error when rendering with Statistics progress bars.
Fixed bug with Quantitation pane where the linear regression divides by zero
Fixed reported errors with FASTA parsing
Changed "Thinning" text to "Condensing"
Fixed bug when passing average fragment mass information to X! Tandem
Fixed bug locking Scaffold to computer
Fixed bug with extracting modifications due to Mascot parsing order rearrangement