SEURAT Version 5.4.3 July 07, 2009
Improvement
- [SEURAT-311]
- SDF Export - Only write out ChemAxon extended smarts (cxsmarts) if
conversion to standard smarts fails
- [SEURAT-312]
- Allow initial search on assay value without providing assay type or
units - e.g. all results < 0.005 regardless of assay name, type or
units
Bug
- [SEURAT-305]
- Need to lowercase username when doing single sign-on.
- [SEURAT-310]
- SEURAT Forms - not clearing fields / constraints properly when move from
compound with data for all form field to compound without data
- [SEURAT-313]
- Batch Cmpd Registration - batch file unload would sometimes fail
(results from more than one file referencing to-be-deleted files cmp, lot
or sample)
SEURAT Version 5.4.2 June 26,
2009
Improvement
- [SEURAT-307]
- Store DBConstants retrieved from the server on a separate thread
immediately after retrieving them.
- [SEURAT-308]
- Too many significant digits in SAR Matrix display and in expanded mode
for Sigma and calculated property columns
Bug
- [SEURAT-306]
- Memory Leak: ClientDbConstantsRepository was holding on to the
DBConstants repository restored from the filesystem when it did not need
to.
-
SEURAT Version 5.4.1 June 22,
2009
Improvement
- [SEURAT-290]
- Large data transfers between SEURAT Server and Client should be
compressed - CWB cache files and large file import jobs are the two most
obvious cases.
- [SEURAT-298]
- Improve SEURAT startup time for systems with very large (> 50,000)
assays
- [SEURAT-299]
- Add button to clear results in Forms view
- [SEURAT-302]
- Improve the time taken to retrieve a large list of compounds into the
Forms interface
- [SEURAT-303]
- Add ability to control the text on the borders around the control panel
in the Form Display
Bug
- [SEURAT-292]
- The presence of corpid_aliases.properties causes SeuratServer to not
start.
- [SEURAT-293]
- Expt Result Type box gets permanently disabled in SEURAT 5.4
- [SEURAT-294]
- If add several assays as add assay columns with same assay name and
different assay types get multiple values
- [SEURAT-295]
- If create job with add rows then Edit->Run job the added row ids get
included in the Corporate IDS constraint -> wrong results.
- [SEURAT-296]
- Corporate ID aliases are cleared when call to updateDBConstants() but
not updated (refreshed/reloaded) resulting in a performance hit for the
next client.
- [SEURAT-300]
- Modify Assay Data by Chemical Lot view to handle legacy data that has
numbers for Expt Batch Number instead of strings.
- [SEURAT-301]
- NullPointerException during start of server if jobs/cwb_cache directory
does not exist
SEURAT Version 5.4 May 26, 2009
- The property syn_search_portal_tab_order
is no longer honored. A new property syn_list_manager_visible_tabs=SEARCH_PORTAL,QUERY_BY_FORM,RESULTS
is now provided to allow customers to decide which tabs (along with their
order) will be seen after login in the List Manager
- The jetty.xml cofiguration
file in the etc subdirectory of the Jetty web server has been modified as
part of the latest release of the SEURAT WebStart client. Customers
upgrading to SEURAT version 5.4 from an earlier version that have made
changes to the jetty.xml should save those changes before upgrading to the
latest SEURAT Webstart Client and reapply those changes after upgrading to
the SEURAT version 5.4
- Customers using the Assay
Data by Chemical Lot View that also have their installation of SEURAT
configured to access their in-house database(s) will need to modify their
mapping file for the syn_property entry for Expt Batch Number to return a
string value rather than a numeric value. This can be done using syntax as
shown below:
{"experiment", "Expt Batch Number",
"#hardcoded1#to_char(customer_table.customer_column)",
"varchar"},\
New Feature
- [SEURAT-266]
- Add EHCACHE as a ClientMoleculeCache option
- [SEURAT-267]
- Add List Manager Options to the Options Dialog
- [SEURAT-269]
- Support MySQL connection strings and add mysql driver jars to
distribution.
- [SEURAT-2] -
Allow constraint on assay type when searching multiple assays
- [SEURAT-255] - Import SD & CWB files via URL
- [SEURAT-280]
- Web Publishing - Publish files to web to view in browser from Assay
Display and Visualization windows
Improvement
- [SEURAT-135]
- Installer do-over for Windows followed by Unix and Mac if time permits
- [SEURAT-256]
- Default to searching primary database if users "databases to
search" preference references old no longer valid database names
- [SEURAT-273]
- Modify SEURAT to allow for Expt Batch Number to be mapped to a database
columns that stores strings (varchar)
- [SEURAT-276]
- Make SEURAT Form population over low bandwidth connections
- [SEURAT-277]
- Improve remote (low bandwidth) startup time when SEURAT mapped to many
database leading to large amounts of DB meta data
- [SEURAT-279]
- Enable a Server Specific idenitifier to be used as the first digit for
jobs running in that servers namespace in order to guarentee there will be
no job id collisions in a multiple SEURAT Server Environment
- [SEURAT-281]
- Allow customers to use properties to determine which style of assay type
"Ki like" or "IC50 like" should be passed on to the
Kinome Tree Viewer
Bug
- [SEURAT-262]
- auth.ldapUser.propertiesRole.xml needs a type qualifier for the
PropertiesRoleDao
- [SEURAT-264]
- bug fix for ignoring sigma tool formulas with no LHS, e.g. a > 10
- [SEURAT-265]
- Fix memory leak of AssayDisplay not being garbage collected when closed.
- [SEURAT-132]
- SeuratServer installer for Windows creates a run.bat with an incorrect
path to JVM if server installed in non-standard location.
- [SEURAT-146]
- Sometime compound structures retrieved from database render aromatized
when then are actually in kekule (non aromatized) form in the DB
- [SEURAT-193]
- Client should download log4j.SeuratClient.properties file from web start
directory (outside of a JAR file)
- [SEURAT-272]
- Sort remains on hidden columns which shouldn't be the case
- [SEURAT-275]
- Sibling tables in Forms dont populate related fields correctly on row
selection. This is due to mixture of constraints applied because of
threaded data population
- [SEURAT-278]
- =all Assay Name logic failing to return correct results when assays are
a mixture of data from different sub mapping files
- [SEURAT-282]
- CWBCacheBuilder (Cache Builder) throws a NullPointerException when a job
is deleted after it gets the list of jobs to process.
- [SEURAT-283]
- List Mananger window's menu icons appear only as checkmarks on Windows
Vista
- [SEURAT-284]
- Convert username to lower case and strip training whitespace to avoid
multiple user folders with mixed case and spaced from being accidentally
created
- [SEURAT-285]
- Change import code so that SD file imports do not identify a column as
the "Compound Structure" column
- [SEURAT-286]
- Stop table cell editing in ResultDisplay_LoadFileDialog prior to doing
import.
- [SEURAT-289]
- If register data from multiple files that reference the same sample then
delete of the data loaded from either of the files fails due to constraint
violation
SEURAT Version 5.3.3 March 6th 2009
Bug
- [SEURAT-257]
- Client molecule caching was broken due to MaxKeySize being set to 0
Improvement
- [SEURAT-252]
– Further work: SEURAT Web Services. SEURAT can now be controlled via a
set of Web Services. These Web Services can be used to:
- Open an existing SEURAT
LiveView
- Open a saved and
named SAR LiveView
- Open a saved and
named SEURAT Visualization Suite LiveView
- Open an SDF or CSV
file so that, for example, SEURAT can be used to present results from the
executing of a Pipeline Pilot protocol.
- NOTE: If the SEURAT
client is already running upon receipt of a web service request the
current instance will handle the request. If the SEURAT client is not
running at the time the Web Service request is made, it will be
automatically launched and then the Web Service request fulfilled.
SEURAT Version 5.3.2 February 27th 2009
PLEASE NOTE: Existing customers upgrading from SEURAT 5.2.x
or earlier will need to update their database(s) to be compatible with
the latest version of SEURAT 5.3.x. Instructions on the steps to
follow to upgrade your database(s) can be found in the “Upgrading your database
tables for use with SEURAT 5.3” guide on the documentation page of our website.
Bug
- [SEURAT-245]
- Permit Substructure match/count columns to be in Scatterplot axis list
drop-down
- [SEURAT-247]
- Histogram does not immediately refresh when new color settings applied
- [SEURAT-250]
- Double-clicking on compound ID to trigger an "All Assays"
search is slow the first time.
- [SEURAT-251]
- "Run Job By Job ID..." Menu Item does not honor cache
settings.
Improvement
- [SEURAT-244]
- Make HeatMap and Trellis plots have tear-off toolbars like other plots
do
- [SEURAT-249]
- Make find bar visible by default for AssayDisplay and ResultDisplay.
Remove "Highlight" button.
- [SEURAT-252]
- On client startup, automatically open job and visualization specified on
command line
SEURAT Version 5.3.1 February 14th 2009
Bug
- [SEURAT-238]
- Sorting sometimes changes which column gets sorted when you add a column
- [SEURAT-240]
- Edit Template form is not setting the correct database when it is not
the default
- [SEURAT-241]
- Template Form =any/all not working in pivot DB mapping configurations
- [SEURAT-243]
- Empty "" rather than null Qualitative value during assay
result registration from a CSV file would cause valid quantitative assay
result to be ignored.
SEURAT Version 5.3 February 10th 2009
Bug
- [SEURAT-35]
- Visualization snapshot does not open correctly if rename a column in
Assay Display
- [SEURAT-98]
- You cannot hide Summing Tool (Formular) columns from the menu option.
- [SEURAT-164]
- Excentric labels broken in new visualization
- [SEURAT-194]
- Visualization snapshot when reopened does not size windows correctly and
visual tab settings for "size by", "color by" and
"label by" are not restored
- [SEURAT-203]
- Histogram Visualization does not produce correct labels when plotting
date columns
- [SEURAT-205]
- Exceptions are not being reported by Compound Whiteboard
SMPCStrucSearchJobModel when it performs a structure search
- [SEURAT-214]
- Serialization Unmarshall Exception prevents job from running
- [SEURAT-218]
- OutOfMemory when combine structure, all lots and all assays search
- [SEURAT-221]
- Add rows in Assay Display fails if underlying data produced from a
combine job operation.
Improvement
- [SEURAT-105]
- Allow for DB mapping files that are used *only* to populate forms.
- [SEURAT-152]
- Allow DB Constants refresh at a specified time each day.
- [SEURAT-170]
- Histogram Coloring: show proportional bands in each bar
- [SEURAT-185]
- Heat Map Header Rotation
- [SEURAT-225]
- Make buttons on Assay Display bigger
- [SEURAT-229]
- Ability to log a comment along with an SMPC compound registration and
retrieve it later.
- [SEURAT-230]
- Add Regression statistics and Error Bars to Scatter Plots
- [SEURAT-231]
- Create Secondary X or Y axis on Scatter Plot
- [SEURAT-232]
- Support localized numbers on plots (e.g. European format)
- [SEURAT-233]
- Provide option to display values in Heat Map cell
New Feature
- [SEURAT-97]
- Copyright year in the Help About dialog should list the current year.
- [SEURAT-172]
- Add legend to Visualizations
- [SEURAT-177]
- Molecule structure callout
- [SEURAT-204]
- Cache Molecules to Client Local Disk
- [SEURAT-207]
- Query By Form
- [SEURAT-208]
- AND logic for many Assays in search
- [SEURAT-209]
- Folderize Live-Computed Properties
- [SEURAT-215]
- Add ability to truly delete (not just hide) columns
- [SEURAT-216]
- Implement Single Sign-on for Windows environments using NTLogin identity
- [SEURAT-217]
- Trellis Plot: add X and Y plots of ones choosing
- [SEURAT-224]
- "Find" feature for cell data in ResultTable and for jumping to
a specified column in the result table.
- [SEURAT-226]
- Scan the crystal structure directory on the server. Client no longer
triggers a scan. Server will rescan at a specified interval.
- [SEURAT-228]
- Update libraries and database scripts for Chemaxon JChem 5.1.4
- [SEURAT-234]
- Provide project progress visualization
- [SEURAT-237]
- Allow the schema.table used by the Compound Whiteboard to be specificed
in the main configuration file allowing customers to put hypothetical
compounds into any schema
Refactor
- [SEURAT-60]
- Remove all System.out.println and System.err.println statements
throughout the code. Replace with appropriate log4j calls.
- [SEURAT-168]
- Visualization: move plot-specific controls from master control panel to
hideable pane connected to the plot
SEURAT Version 5.2.2 November 6th 2008
Features added and Improved in SEURAT 5.2.2
- SEURAT-191
- Make the ViewSelectionPanel for choosing to display an Assay Display or
a Form Layout more configurable
- SEURAT-197
- Provide option for seurat client to download its log4j.SEURATClient.properties
files from a server
- SEURAT-200
- Rename and reorder Excel Export Menu items
Bug Fixes in SEURAT 5.2.2
- SEURAT-198
- AssayDisplay and ResultDispay copy is not copying empty cells.
- SEURAT-199
- Copy Options dialog brings the wrong window AssayDisplay or ResultDisplay
to the front.
- SEURAT-196
- SDFileToDBConverter appends the SEURAT schema name to JCHEMPROPERTIES
and SYN_STRUCTURE table names when using and Oracle database.
- SEURAT-194
- Visualization snapshot when reopened does not size windows correctly and
visual tab settings for "size by", "color by" and
"label by" are not restored.
- SEURAT-148
- When label by is set to Compound Structure in the visualization package
then the compound structure should be drawn on mouse over.
- SEURAT-202
- Store Chemaxon MarvinSketch configuration.
SEURAT Version 5.2 October 8th 2008
IMPORTANT NOTES ABOUT UPGRADING TO SEURAT 5.2.x from SEURAT 5.1 or earlier
Please read and follow the instructions given in this
section immediately after upgrading to SEURAT 5.2 but before you
try to run SEURAT 5.2.
Update your Compound Whiteboard Tables
SEURAT 5.2 requires changes to the tables that hold the
compound structure definitions stored in the database tables that support the
Compound Whiteboard functionality in SEURAT.
In order to update these table definitions you need to do
the following:
- Open
a cmd window (Windows Installations) or a unix window and change directory
into the SEURATServer directory of the SEURAT Server installation.
- Issue
the command
- java –cp seurat-server.jar chemaxon.jchem.JchemManager
- At
the connection dialog presented enter the following information
For ORACLE
|
JDBC Driver
|
oracle.jdbc.driver.OracleDriver
|
|
URL of database
|
jdbc:oracle:thin:@//localhost:1521/xe
or
jdbc:oracle:thin*:*@ localhost:1521:sidname
|
|
Property Table
|
jchemproperties
|
|
Login name
|
seurat
|
|
Password
|
seurat (usually)
|
For PostgreSQL
|
JDBC Driver
|
org.postgresql.Driver
|
|
URL of database
|
jdbc:postgresql://localhost:3247/synaptic
|
|
Property Table
|
jchemproperties
|
|
Login name
|
postgres
|
|
Password
|
<your password> (chosen during install)
|
- You
will be asked if you would to update your table definitions. Choose yes.
- Once
updated you will be asked if you want to update the data in those tables.
Choose yes.
This will bring you database table definitions in-line with
the latest version of SEURAT. If you experience any difficulties in upgrading
your table definitions please contact Synaptic Science at support@synapticscience.com or on
301 915 0274.
Features added and improved in SEURAT 5.2
- [SEURAT-21]
- No way to easily tell num cmps selected in visualizations
- [SEURAT-154]
- Ability to specify that the forms view should be the first view to be
shown, rather than the grid view.
- [SEURAT-173]
- The Copy Options dialog should not be displayed if just copying the
corporate id column.
- [SEURAT-175]
- Display Low Memory warnings when the SEURAT Client is running out of
memory.
- [SEURAT-178]
- Implement mechanism to push Assay Display "Live View" data
into Spotfire as complement to SEURAT visualization suite.
- [SEURAT-182]
- Modify the startup scripts to support the new ChemAxon 5.0 licensing
(use CHEMAXON_LICENSE_URL property)
- [SEURAT-16]
- Greater add column flexibility to return multiple columns, URLS,
molecules
- [SEURAT-66]
- Sharing Heat Map Settings
- [SEURAT-77]
- Expand kinase panels handled by Kinome Tree Viewer to include Invitrogen
data
- [SEURAT-90]
- Desktop icon not getting installed during WebStart deployment of SEURAT
- [SEURAT-93]
- Make sure primary db name appears first in lists of Assays in
AssaySelectionDialog
- [SEURAT-94]
- Present sub folders for assay categorization in alphabetical order
- [SEURAT-116]
- Remove BuildInfo class. Replace with Version class. Version can be
pulled from JAR MANIFEST.MF
- [SEURAT-117]
- Update Eclipse build and Ant built to use new Log4J log4j.1.2.15
- [SEURAT-130]
- PropertiesUserDao should reload its user data if the timestampl of the
properties file changes.
- [SEURAT-137]
- Add a Histogram Bar Chart visualization to the SEURAT visualization
suite
- [SEURAT-149]
- Closing the main SEURAT visualization suite window should cause the
Control Panel (if open) to close as well.
- [SEURAT-160]
- Enable copying Job Name from the Update Template dialog
- [SEURAT-163]
- HeatMap Image Capture: capture all rows, including those not displayed
- [SEURAT-169]
- Histogram selection rendering
- [SEURAT-174]
- Histogram dynamic filtering
- [SEURAT-179]
- Remove implicit Lot Number column from Visualizations
- [SEURAT-180]
- Ability to choose ChemDraw as the default drawing tool. Requirement
changed to updrade to Marvin 5.0, since skins appear to be a sufficient
fix.
- [SEURAT-181]
- The Expand Box should have the full names in the pulldown...could do
something like: L: Lot, D: Date, P:
Protocol
- [SEURAT-186]
- Put Histogram Labels below Axis
Bug fixes in 5.2
- [SEURAT-96]
- DnD move of Column from far right in AssayDisplay causes auto-scroll to
beggining (1st page) of JTable
- [SEURAT-125]
- "Save template" menu item brings up the Update Template dialog
with the Title disabled.
- [SEURAT-134]
- Cannot paste into the new Chemaxon 5.0 MViewPane in the TemplateForm
StructureSearch panel
- [SEURAT-138]
- Cant resize properties pane (top LHS) of Compound Whiteboard when using
JChem 5.0.5 MarvinSketch in ChemDraw mode.
- [SEURAT-139]
- JChem 5.05 - You have to click twice in the compound structure to be
able to copy a compound structure. You should only have to click once.
- [SEURAT-140]
- JChem 5.0.5 - When you select, a compound in the compound structure
column, the border doesn't appear. It will appear if you resize the
column.
- [SEURAT-153]
- Long assay names mean cant see structure sketch area in SAR Tool
- [SEURAT-155]
- SARSnaphot logic for saving a new SARSnapshot from an existing
SARSnapshot results in overwriting original SARSnapshot data.
- [SEURAT-156]
- SAR Table only checks to see if you are going to overwrite the current
SARSnapshot name. It does not warn if you are going to overwrite any other
SARSnapshot already saved.
- [SEURAT-159]
- Show Details in New Visualizations does not work if done after a scatter
plot has been opened
- [SEURAT-161]
- Cannot delete a job bookmark for a job that no longer exists (deletion
of bookmark is not being saved)
- [SEURAT-162]
- Need to lock files while exporting data
- [SEURAT-166]
- Trellis Plot: axes missing for some plots
- [SEURAT-176]
- When you paste into structure panel in Template Form, the TemplateForm
is not recognizing that there is a structure in there.
- [SEURAT-183]
- IllegalStateMonitorException in DatabasesMetadata
- [SEURAT-184]
- Dupes when adding a corporate ID row and adding rows in Compound
Whiteboard (aka SMPC)
SEURAT Version 5.1 August 11th 2008
Features added and improved in SEURAT 5.1
- [SEURAT-110]
- Need to disable the e-mail job menu itemwith a property in the primary
mappings file.
- [SEURAT-44]
- Unable to specify color for range outside existing data values
- [SEURAT-102]
- Remove automatic column sort of the compound id
- [SEURAT-10]
- Ability to control what gets seen when click expand button
- [SEURAT-22]
- Ability to control how heat map AND assay display columns are colored
- [SEURAT-37]
- Improve connection pooling so we can more easily set a maximum number of
database connections used by SEURAT
- [SEURAT-68]
- Have an "Are you sure you want to exit SEURAT?" dialog when
closing ListManager
- [SEURAT-69]
- Need to be able to gray out menu items that lead to tools (e.g. SMPC,
ASC, GVK) that are not currently configured to work.
- [SEURAT-83]
- Add ability to specify default and per user overridable max structs
returned from a struct search
- NOTE: Existing
customers that have SEURAT configured to use an Oracle cartridge for
structure searching via a call to a script will need to modify that
script to expect an additional option, -n that will specify the maximum
number of compounds to return. This value should then be passed to the
cartridge call or added using rownum < max logic to the SQL generated
in the script.
- [SEURAT-32]
- Ability to aggregate exp results in config file per assay type and/or
column name.
- [SEURAT-56]
- Add DB Columns would present duplicate values for Corporate ID entries
in syn_add_cols if retrieved with Sample ID entries AND many way joins
produced slow performance
- [SEURAT-62]
- Allow specification of Never, Nightly and When Opened as the cache
update policy for a job
- [SEURAT-63]
- Better handling of multiple DB's (DB writeback has its own menu, can
choose DB to write to via GUI, auto filter to correct DBs during search)
- [SEURAT-64]
- Ability to organize the Database columns into folders, as can already be
done with the Assays
- [SEURAT-67]
- Maximize window button on "Assay Selection Dialog Box" or just
increase default window size
- [SEURAT-70]
- Optionally remove the 'Include Prefix' checkbox in favor of auto-completing
prefix when searching primary db and expecting explicit IDS otherwise
- [SEURAT-78]
- Ability to save SAR snapshots (and matrix representations), just like
plots.
- [SEURAT-80]
- Support copy to clipboard from any SEURAT visualization
- [SEURAT-81]
- Add the ability to set the background color any SEURAT visualization
- [SEURAT-85]
- Remove '+' and '++' characters when exporting to excel or csv.
- [SEURAT-89]
- Fix sorting in ResultTable and ResultTableModel to be cumulative when
multiple columns are selected and to revert to default sort order when
cleared
- [SEURAT-95]
- Change logic for sorting AssayValues in an Assay Result to allow for
mixture of quantitative and qualitative data for an assay
- [SEURAT-84]
- Load results from CWB cache should be performed on a thread
- [SEURAT-119]
- Allow searching on alias names wherever Corporate ID's can currently be
supplied
- [SEURAT-123]
- Add ability to introduce Oracle hints to queries generated by add column
operations
- [SEURAT-124]
- Automatically sort the assay protocols when presented in the pick list
of Template Form and its associated Protocol Selection Dialog.
- [SEURAT-108]
- The PropertiesRoleDao needs the concept of default roles.
Bug Fixes in SEURAT 5.1
- [SEURAT-4] -
Cant delete SMPC compounds
- [SEURAT-24]
- Image columns get exported to excel as underlying numeric assay value
- [SEURAT-72]
- Should not have hyperlinks on sub lot entries of a row in the Assay Data
by Chemical Lot display
- [SEURAT-91]
- Sort order is lost (or reset) when you add a Property
- [SEURAT-122]
- Cant delete a bookmark that referes to a non existent job, change to
allow deletion of "orphaned bookmarks"
- [SEURAT-126]
- Job Table doesn't update the first time you click on another users
folder in the "Other Users" folder in the bookmarks.
- [SEURAT-127]
- Cache file is always being retrieved when you view results regardless of
useCacheIfAvail flag.
- [SEURAT-107]
- Database menu item not enabling for Administrator role
- [SEURAT-114]
- Structure search in SMPC always reports compound with CRX- prefix
ignoring settings in corporateid.properties
- [SEURAT-103]
- Max Compounds setting is the Advanced Search is not being honored
- [SEURAT-104]
- exclude operator setting is not restored properly
- [SEURAT-106]
- Max Compounds setting is not honored in SMPC (now Compound Whiteboard)
Jobs.
- [SEURAT-33]
- Launch from job URL hangs client on Windows XP
- [SEURAT-49]
- Inconsistent bookmark versus job title (TRAC 61 in GNE's bug tracking
system)
- [SEURAT-55]
- Compound registration detects cmpds with diff stereochemistry as the
same compound and fails to process correctly
- [SEURAT-61]
- Need to be able to rename SEURAT-calculated properties
- [SEURAT-76]
- Conversion of Structure to Smiles leaves out explicit hydrogens
SUPPORT-110
- [SEURAT-82]
- Need to modify seurat.py to use REMOTE_ADDR instead of REMOTE_HOST
- [SEURAT-86]
- New Search Scientific Notation Bug
- [SEURAT-92]
- Expr Date (Experiment Date) operator in advanced search is disabled
after selecting a date.
SEURAT Version 5.0 April 6th 2008
NOTE: Version 5.0 of SEURAT
requires a 1.5 JVM or greater to run. If you want to run under a Java 1.4 JVM
then you will have to use version 4.5.1 or earlier.
Features added and improved in SEURAT 5.0
- SEURAT
now supports single sign-on. Customers can now configure SEURAT to authenticate
users via any LDAP directory (e.g. Active Directory). See the SEURAT
administrators guide for details of the configuration file entries
required to enable this feature.
- NOTE:
The scheme prior to version 5.0 of using the underlying databases
authentication is still available and remains the default for backward
compatibility.
- The SEURAT
Server can now be configured to periodically refresh its database
constants. Database constants form the basis around which pick lists are
built by the SEURAT Client upon each login.
- By
default periodic update is turned off
- To
enable periodic update of database constants you must set the System
Property “db.constants.refresh.interval” and set its value to the number
of minutes that should pass between each refresh.
- An
example setting for refresh every hour would be -Ddb.constants.refresh.interval=60
- Two
new features for organizing/clustering a sparse table of data have been
added to the Assay Display. These new options are found under the
Edit->Organize table option. The three sub options are:
- Data
existence
i.
A fast way to bring the most assayed
compound to the top and then the most often run assays to the left so that all
“interesting” data is closer to or in the visible region of the table
- Euclidean
distance->Automatic
i.
Will compare all quantitative columns
using a nearest neighbors calculation based on Euclidean distance
ii.
NOTE: The Euclidean distance calculations
are slower than the data existence method.
- Euclidean
distance->Manual
i.
Same as above but only for the subset of
columns you choose in the dialog presented when this option is selected
- SEURAT
can now be configured to send bug reports from the Server rather than
client to get around environments where Java Mail send from the client is
blocked. Two new properties were introduced to support the server send of
bug reports and feature requests:
- syn_bug_report_sender
which can be either “client” or “server”
- syn_bug_report_server_script
which is syn_bug_report_sender is “server” gives the name of the script
that should be called to effect the sending of email.
- The
script will be called with –s <subject> <report_file_name>
- Tables
in SEURAT forms are now sortable by a single click on a column header.
Sorting on multiple columns at the same time is possible to break sorting ties
in the preceding column.
- Tables
in SEURAT Forms can now be defined to show only a subset of assay data by
providing a list of partial or full assay name matches in the second row
of the table model painted via NetBeans (or another editor). Entries like
PK_%, %TRFRET% would populate a table only with assays that had a name
starting with PK_ or had the term TRFRET somewhere in the assay name.
- SEURAT
forms now support two or more tables at the same level and subject in a
form in case you want to split very large (number of columns) tables into
2 or more tables to avoid users needing to scroll.
- New
administrator tools are now provided that allow the SEURAT administrator
to run a set of jobs by their job numbers and to be presented with a GUI
showing for each job
- Job
ID
- Owner
- Title
- Description
- Path
to job in job bookmark outline
- The
script to run the admin tool for running an inspecting jobs by job id are
in the SEURATClient directory (C:\Program Files\SynapticScience\SEURATClient
by default)
- A
new administration script is provided in the SEURATServer/bin directory
for parsing cache builder log files for details on which jobs failed and
how long re-cache of each job took. The script is called genCacheBuilderReport.py
and is called passing two arguments:
- Arg
1 = path to a cache builder file
- Arg
2 = name of the file to which output of the script should be written
- The
output of the script is written to the file designated in Arg 2 where the
CSV file produced contains a header row of:
i.
“Job ID,Seconds To Cache,Megabytes To
Cache,Success/Error?”
- It is now possible to add an “All Assays”
add column operation to an existing SEURAT job. This makes it easier to
combine a cutoff query (e.g. Ki < 0.05 uM) with a post addition of all
assay data such that running the job in the future will always find all
new assay results for those compound matching the initial search criteria.
- The SEURAT memory meter has been removed
from the List Manager page as it had caused many users to think that SEURAT
was still loading. The memory meter is now only shown if the property
seurat.test is set to true.
- A
new menu option for running a job by its id has been added to the View
menu of the ListManager.
- Customers
can now set the order of buttons on the toolbar of the ListManager. They
can also remove any of the buttons decide where the vertical separators
should be placed that group toolbar buttons or remove the toolbar
altogether.
- Customers
can now define their own search portal page to be used in conjunction with
the SEURAT main List Manager window. The Search Portal page is designed by
customers using any WYSIWYG GUI builder tool for java (like for example
NetBeans) to paint a JPanel that also implements the Synaptic Science supplied
SearchPortal interface. The SearchPortal interface allows the search
portal author to tell SEURAT which panel should contain the compound
structure for search, which should accept a list of corporate ids.
- Users
can now double click on the compound structure pane in the Structure tab
of the Advanced Search (aka Template Form) window to access the structure
sketching view.
- A
new assay search can now be launched from within the Assay Display based
on the current selection in the table from a right mouse click over any
table cell and selecting the “New Assay Search (all selected columns)
option.
Bug Fixes in SEURAT 5.0
- CWB
Cache builder now correctly deletes the cache results for any job that
fails to run to completion during the cache rebuild process. Without this
fix the next time a user tried to run the job that failed from cache they
would see no results and would have to remember to run from scratch
manually (right mouse option) to see their latest results.
- Jobs
that included Add Property Columns for properties that no longer existed
in the DB Mappings file would cause the cache builder to fail to complete
often tying up large amounts of DB resources. Now even if such jobs exist
and fail during the cache rebuild process, they will not stop the cache
builder from completing cleanly and releasing its resources.
- SEURAT
now properly ignores “bad” or null compounds and continues to render any
in a hit list / analysis set that are valid.
- Rendering
of the SEURAT GUI would sometimes become slow and sporadic after closing a
heat map in SEURAT visualization suite. This issue has been corrected.
- SDF
export from windows used to only put a line feed at the end of each line
rather than the carriage return and line feed windows / DOS expects. This
has been corrected such that SEURAT will output the appropriate end of
line characters for each operating system.
- SEURAT
client would often hang when launched from a URL on dual and quad core
machines due to some threading issues. This has been resolved.
- Very
long assay names used to cause the Advanced Search window to have
scrollbars obscuring the magic wand button which gives access to the Assay
Selection Dialog. SEURAT not employs stepped combo boxes to avoid the
scrollbar appearing for very long assay names.
- Add column
operations would fail (in 4.5.1 version of SEURAT only) if there existed
two or more assay names where one assay name was a subset of another
longer name. This has been corrected.
SEURAT Version 4.5.1 February 25th 2008
Bug Fixes in SEURAT 4.5.1
- Save of Assay Display sometimes fails
when prompted to save after close window: after changes have been made
in the Assay Display and a user clicks on the Windows close button (top
right hand corner) they are prompted to save their work. Sometimes when
saving just before closing the window the cached copy of the results would
not get written out properly. This could have the following effects from
the users perspective:
- When
re-run from cache none of the updates for the job can be seen
- A
visualization snapshot created before the save is no longer available
from the Tools->Visualizations menu.
- Improved memory management and
instrumentation in SEURAT Server: Installations of SEURAT with
thousands of SEURAT jobs, especially when many of those jobs represent the
results of loading large SDF and/or CSV files into SEURAT to share across
projects, were starting to consume a lot of memory. Now the SEURAT Server
can support many more jobs in a much smaller memory footprint and has been
instrumented to give additional information as to which jobs are causing
the greatest memory consumption.
SEURAT Version 4.5 February 18th 2008
Features added and improved in SEURAT 4.5
- 3x to 5x speed improvement for
retrieval (from database) and rendering of compounds
- Assay Display cut and paste enabled (e.g.
between SEURAT and Excel): Cut and paste operations now work within
the Assay Display. This allows for cut and paste between the SEURAT Assay
Display and Microsoft Excel amongst other applications. Data can be pasted
directly into SEURAT user defined columns from Excel, other applications and
from any of the other columns in the AssayDisplay itself.
- User
defined check box columns will accept integer values (0 = false, non zero
= true) values as well as the words true/false during a paste operation
- Right mouse on column header to hide
column: It is now possible to hide a column by selecting the “Hide
Column” menu item from a right mouse click with the mouse over the column
header of any column in the Assay Display within SEURAT.
- Java 1.6 compatibility: SEURAT can
now be run under Java 1.6 as well as Java 1.5 and 1.4
- Heat Map Coloring can be controlled on
a per Assay Type basis: The SEURAT Heat Map now supports the ability
to designate the data range, scaling (linear, exponential, e to the X and
natural log) and hot and cold colors the heat map will use when coloring
assay results on an assay type by assay type basis (as well as per column).
These options can be accessed and controlled via the right mouse click
“Heat Map Color Settings” menu with the mouse over any of the column
headings of the heat map.
- Can now specify JChemBase structure
and properties table names: Two new properties have been added to
support the use of JChemBase structure searching that allow the customer
to tell SEURAT the name of their JchemBase structure and properties
tables. Those properties and their default values (suitable for use with
the Synaptic Science native schema) are:
- syn_struct_search_jchembase_table=SYN_STRUCTURE
- syn_struct_search_jchembase_prop=JCHEMPROPERTIES
- New property to allow support for overlay
images in AssayDisplay: A new property syn_use_overlay_url has been
introduced that when set equal to “true” will cause SEURAT to call a
single image producing url passing the assay results as a comma separated
argument to that URL. When syn_use_overlay_url is set to “false” or
undefined then multiple urls, one for each assay result will be called and
a table of images displayed within a single cell of the Assay Display.
- Improved logic when operators excluded
from aggregate results across lots in the Assay Display: Now when the
user requests to exclude assay results with operators from the calculation
of aggregate values (e.g. arithmetic mean, geometric mean, min, max) in
the Assay Display the following logic applies:
- If
all the values have operators then all those values are included so that
at least the user knows assays have been run. The value is followed by
“++” to show two or more values have operators.
- If
assay results are a mix of values qualified by operators and values that
are not qualified the aggregate value is calculated using only the values
without operators. Mouse over however shows the complete range of values
so the user is aware that qualified values with operators exist.
- NOTE: This change in rendering logic means that for any existing cached
job results where the column contained only qualified values and the
aggregation policy was to exclude values with operators then the cell
will now display >null (single value) or null++ (multiple values all
qualified) until the job is re-executed from scratch.
- Excel converter host and port can now
be specified through system properties: In addition to being able to
set the excel converted hostname and port number through the “Prefs…”
button on the login screen the SEURAT administrator can centrally override
and change these properties by specify the excel.host and excel.port
system properties in the resources section of the SEURAT WebStart client
JNLP file. The entries would look like:
- <property
name="excel.host" value="somehostname"/>
- <property
name="excel.port" value="80"/>
- First four Quick Searches can now be
edited using the edit job button: Previously an attempt to edit a
saved job that was the results of running one of the first for quick search’s
(all results for a corporate id(s), prep data for corporate id(s),
background data for corporate id(s), all results for an assay) generated a
dialog indicating editing was not supported for quick search jobs. Now
users can edit jobs created using the first four quick search options.
- New properties and command line args
for controlling the Cache Rebuilder:
- The database username and password
can now be passed as arguments 2 and 3 to the cache re-builder
(previously it always ran as the special user SEURAT).
- In
addition a new system property cache.dontupdate.regression if set to true
instructs the cache re-builder not to update jobs under the “Regression
Testing” jobs folder of the user under which the cache builder is run.
- Redesign of the Combine Jobs Form: The
form used to invoke a combine job has been redesigned to be more
intuitive.
Bug Fixes in SEURAT 4.5
- First digit of number of molecules
obscured in status bar: in some circumstances the first digit of the
number of molecules reported in the status bar of the AssayDisplay would
be obscured making it seem like less compounds were present in the display
then was actually the case. This has been corrected.
- Search for “All Assays” during given
time period not working: A Template Form (Advanced Search) specifying
“All Assays” and an experiment date constraint would generate a dialog box
saying insufficient constraints have been set. This has been corrected and
now the search runs as expected.
- Search for “All Assays” combined with
a Structure Search slow: Searches that combined a request for all
assay results against a set of compounds identified by a structure search
was extremely slow and would sometimes cause the SEURAT Client (and very
occasionally SEURAT Server) to run out of memory. This has been corrected
such that the query runs much more efficiently and the out of memory
issues have been resolved.
- Assay counts not working when
multiple/sub mapped databases used: When syn_show_counts_per_assay=true
SEURAT provides a count of the number of results available in the
database(s) for each assay. This was not working correctly when SEURAT was
configured to access more than one database or a single database multiple
ways (sub mappings). This has been corrected.
- Search for quantitative assay value
cutoff combined with lot constraint still returns all lots: When a
search involving a cutoff on a quantitative assay value was combined with
a request for only a particular lots worth of data SEURAT would return all
lots. Now SEURAT correctly returns only those values that are both for the
lot specified and pass the quantitative cutoff constraints.
- Add assay column operations fail for
assays including ‘(‘ in name/type: Some naming conventions used by
customers that made use of the ‘(‘ and ‘)’ characters in either the assay
name, assay type or both would cause add assay column operations to fail
due to parsing errors. This has been corrected and customers are now free
to name their assays and assay types without restriction.
- Add property and user defined columns
to file loads with auto id: SEURAT used to only create the Result
Display for a file loaded into a job that was using auto identifier
generation supplied by SEURAT. Now opening a file into SEURAT using auto
identifiers correctly creates the Assay Display from which property
calculations and user defined columns can be added to augment the file
data.
- Add database columns to file loads
with corporate id: Previously could not add database columns to the
results of loading a file (SDF or CSV) into a job when SEURAT was
configured to access multiple/sub-mapping databases unless syn_add_columns
and syn_properties were identical in all mapping files. This has been
corrected and in addition in a multiple/sub-mapping database environment
is much more efficient.
SEURAT Version 4.4 January 2nd 2008
Features added and improved in SEURAT 4.4
- Improved support for forms: Forms
can now be used to extend SEURATs ability to access additional data from
your federated data sources.
- Added
example In Vitro Summary (averaged) results and In-Vitro Raw (including
Well information) results forms as a means of showing how SEURAT can
support master detail tables in forms built using the SEURAT form builder
mechanism to any level.
- Form
designs can now specify multiple join conditions and the order of the
joins to be applied when SEURAT reacts to selection events in a master
table within a form to populate a subordinate table.
- Forms
now appear in a Tabbed Pane layout to make it easier to switch from one
form to another. The “current molecule” and molecule sort order is
synchronized between forms and first and last molecule navigation buttons
have been added.
- Geometric mean / Improved Averaging
Control: Customers can now configure SEURAT to calculate the geometric
mean of experimental results across lots and protocols in addition to the
already supported arithmetic mean, median, max and min functions. In
addition:
- Can
configure at the global level which aggregate function to use by default
and override on an assay by assay basis
- Can
configure at the global level whether values with operators should be
included in aggregate calculations or not and override on an assay by
assay basis.
- Improved PyMol integration: It is
now possible to load information about pdb and pse files into the Synaptic
Science schema from sdf and csv files that includes atom display and atom
selection paths which can then be use to incrementally overlay multiple
objects into a single PyMol session using the first Protein sent into
PyMol as the frame of reference.
- It is now possible to search on
multiple protocols at once: The SEURAT Template Form search GUI now
allows the designation of multiple Assay Protocols as search criteria in
any initial search.
- Display assay and/or assay / assay
type counts: SEURAT can be configured to show the number of compounds
that have been run in a give assay and/or assay and assay type combination
whenever those assays are presented for selection during search
construction. This makes it clearer where the holes are in current assay
data.
- Greater flexibility when displaying
images in Assay Display: Customers can now choose to configure SEURAT
to interpret as images any assay result using a much wider range of match
techniques those being indexOf, lastIndexOf, equals, equalsIgnoreCase,
startsWith and endsWith
- Added ability to not update cache
files: Customers requested a way to exclude certain SEURAT jobs from
the overnight cache rebuild process so that the jobs “snapshot in time”
cache could be used to compare these results to a job that retrieves the
most recent data for the same constraints.
- Similarity sort re-enabled: Similarity
sort based on similarity to the currently selected compound in the Assay
Display has been re-enabled either using SEURATs own native tanimoto
calculation or via a script provided by the customer.
- SAR Analysis and Matrices Work on
Compound with Salts: it is now possible to perform an SAR Analysis on
compounds including salts.
- Can now control who bug / feature
reports are sent to: it is possible to designate who will be the
recipient of bug and feature requests submitted through the SEURAT bug
report window. In addition customers can configure through which SMTP
server, messages should be sent.
- The link followed from the main Help
menu is configurable: by default SEURAT takes you to the on line
version of the SEURAT users guide from the help menu. This can now be overridden
by each customer to send a help request to any web addressable URL.
- Easier license file management:
License checks are now performed on the server rather than on the client.
This makes deployment of updated or new licenses easier.
- Globally control formatting of assay
results for display: it is now possible to specify via SEURAT
configuration settings (using the syn_df0… syn_df10 properties) the number
of decimal places that should be displayed for quantitative assay values.
Different formatting rules can be applied depending on the magnitude of
the assay result.
Bug Fixes in SEURAT 4.4
- New assays missing after upload file
data to DB: Occasionally new assays introduced during an upload of
data from an SDF or CSV file into the database did not appear in the drop
down lists of the SEURAT GUI until after a SEURAT Server restart. Now the
new assays appear after the file upload.
- Add rows not calculating column values
when run on results from Quick Search: Add rows operations now work
consistently regardless of whether the job to which rows are being added
was created using Quick Search.
- All rows hidden when job with some
hidden rows is duplicated: Sometimes when a job that was saved with
hidden rows was duplicated and re-executed the new copy of the job would
mistakenly end up with all of its rows hidden. This has been corrected.
- Add rows to job with saved hidden rows
not working: there were several situations where adding rows to a job
that already had hidden rows would result in incorrect behavior (too many
rows hidden, added row not showing up because it was itself hidden). These
problems have been corrected.
- Pasted structure for search only
recognized after open editor: Even though it has always been possible
to paste a structure directly into the structure tab of the Template Form
the structure was not recognized as a search parameter until opened into
the compound sketching window (via the magic wand button). This has been
corrected such that all that is needed is a paste into the structure tab
for a structure search.
- Unable to clear lot date registered
and prepared constraints once set: now clicking on the Clear All Tabs
or Clear This Tab buttons in the Template Form correctly clears the Lot
Date Prepared and Lot Date Registered fields if set.
- Units not displayed in headers until
an Add Data operation performed: now the units display in the header
correctly at all times without the need for an add data operation.
- Copy and paste of structures into
office documents: In version 4.3 some jars were missing that allowed
the correct copy and paste semantics between SEURAT and office products
such that the paste operation would fail. This has been corrected in
version 4.4.
- Hidden columns folder sometimes
missing: for jobs with hidden columns sometimes the hidden columns
folder of the add data dialog did not appear to allow the hidden columns
to be added back to the job. This has been corrected.
- Improved memory management when
running CWB Cache Rebuilder: The molecule caching mechanism could on
occasion exhaust the SEURAT Servers memory or leave it will too small an
amount of memory after a CWB Cache rebuild. Now a new system property max.cached.molecules
has been introduced that can be passed to the CWB Cache Reuilder scripts
java command to set the maximum number of molecules that can build in the
cache before the cache is flushed. If the system property is not passed to
the cache rebuilder the default maximum is 500 compounds.
SEURAT Version 4.3 September 10th 2007
Features added and improved in SEURAT 4.3
- Support for forms added: SEURAT
now supports a form based view of compound data. A sample “Chemistry data”
form has been supplied in this release as an example of the kind of forms
that can be created for use within SEURAT. In the next SEURAT release
users will be able to define their own forms covering any combination of
chemical, biological and preclinical data already accessible through SEURAT.
- SEURAT
forms are accessed through the View menu of the Assay Display created
from the execution of a SEURAT process pipeline (job).
- The
form names and the class that gets instantiated to provide the form
content and layout are controlled via properties in the
pri_mappings.properties file of the SEURAT Server.
- Better performance and user
notification regarding failover: SEURAT’s failover mechanism now
informs the user when they have been switched to a failover or backup
server as well as when they have been switched back to the primary SEURAT
server. Customers had requested this feature to ensure users were at least
aware that they were being routed to a backup server in case that server
was less performant (due to inferior hardware resources) or had the
potential to be slightly out of date (due to customers own policies for SEURAT
job replication) in terms of the serving of SEURAT process pipeline (job)
information.
In addition failover logic is now
only activated when the property backup.server.url is defined so that in the
case where no failover server is specified there is no performance penalty for
“keep alive” pings to the SEURAT server.
- New debug options to make mapping SEURAT
to Databases easier: two new properties syn_query_builder_debug and syn_query_comp_struct_debug
were added to allow customers to request that SEURAT output a high level
of debug information during testing of a new mapping file (e.g.
pri_mappings.properties for a third party database).
- Allow use of different database for compound
retrieval: a new property syn_cmpd_diff_database (along with 7
additional related properties) allows customers to specify that compound
structure data be retrieved from a different database than that of the
database containing chemical, biological and preclinical data related to a
compound. This increases the degree of flexibility that SEURAT can support
when mapping its query, analysis and reporting functions to work against data
residing in existing databases.
- Register biological data against compounds
registered external to SEURAT: a new property syn_external_struct_registration
allows customers who use an application other than SEURAT to register
their compound structures to still use SEURAT to register lot, protocol,
biological assay, preclinical and property data against those same
compounds using the SEURAT “load file data into database” capability.
- Compound retrieval, averaging and job
save performance improved: when dealing with a relatively large number
of compounds (> 4000) in the Assay Display SEURAT now calculates
average values, loads compound structures and saves a copy of any process
pipeline results in the background allowing customers to continue working
on other tasks.
Bug Fixes in SEURAT 4.3
- Incorrect constraints on Add Assay Column:
when the new property syn_sce_assay_queries was set to “true”
incorrect constraints were generated when adding assay columns to the
Assay Display resulting in longer query times than were necessary. This
has been resolved.
- Duplicate Observation Types (aka Assay
Types) Generated: under some “load file data into database” situations
duplicate observation type entries were generated into the Synaptic
Science native database. This issue has been corrected and a commands
supplied in the updateSchema.sql script (in the postgresql or oracle SEURAT
Server sub directories), which SEURAT administrators need to run after
each upgrade, to remove the duplicates.
- Unable to unload data loaded into
database: when more than one file was loaded that referenced the same
protocol number of a given assay name there was a bug that would stop the
unload of any of these related files. This has been corrected such that
now any of the files that share the common observation protocol can be
unloaded to completely reverse a “load file data into database” operation.
SEURAT Version 4.2.3 July 27th 2007
Features added and improved in SEURAT 4.2.3
- Java 1.5 compatibility; SEURAT can
now be run under Java 1.4 or Java 1.5 JRE’s.
- Easier to configure database
connection settings; Database connection settings which you could set
centrally to seed client first login settings are now honored each time a
user logs in. This means that a SEURAT administrator can change database
connection settings centrally in the pri_mappings.properties file (or
equivalent) and have these automatically propagated to all client users.
- Database mapping configuration file
syntax extended to support hardcoded values and data pivoting: Now
when a SEURAT administrator is configuring SEURAT to access an existing
database they can make use of the #hardcodedN# and #pivotedN# statements
within the pri_mappings.properties file to tell SEURAT that a literal
value should be returned or column values pivoted into rows respectively.
More details can be found in the comments section of the latest
pri_mappings.properties.example file in the SEURAT Server installation
directory.
- Database mapping performance improved:
When database values are split across a large number of tables SEURAT can
be instructed using the new syn_sce_assay_queries property in the
pri_mappings.properties file to only search tables known to contain
particular assays by name when a query is constrained by assay.
Bug Fixes in SEURAT 4.2.3
- List manager table now sorts and
refreshes correctly when click on column headers; The initial List
Manager table can now be sorted by single clicking in the column headers
in the same manner as sorting is supported in the Assay Display.
- Importing from CSV last column
exclusion; There was a bug that stopped imports from a CSV file into
the database from working when the user elected to exclude the last column
in the CSV file from the import. This issue has been resolved.
NOTE: Due to
differences between Java 1.4 and Java 1.5 if you have an existing installation
of SEURAT running under Java 1.4, plan to move to run under Java 1.5 and have SEURAT
jobs involving the use of User Defined Scaffold Columns then you need to do the
following when upgrading:
- Install version 4.2.3
- Run the new version (client and
server) initially still under Java 1.4
- Run the CWB cache rebuilder or
manually run and then resave all jobs that include User Defined scaffold
columns.
- Once this has been done you can then
start running the client and server under Java 1.5
SEURAT Version 4.2.2 May 24th 2007
Features added and improved in SEURAT 4.2.2
- Background loading of images in Assay
Display; SEURAT will now load images (for things like IC50 thumbnails
and compound annotation information) in the background leading to
significant performance improvements.
- Add Column Names can be changed;
The name of the database columns and property columns that can be added as
new columns in the Assay Display can now be changed without breaking
existing jobs that reference the old deprecated names for those same
database columns and properties.
- Better color separation when coloring
on categories in visualizations: When coloring by a “categorical”
column (like for example scaffold class) in SEURAT visualizations, better
color separation is achieved to make each category more visually distinct
from every other category or cluster.
- Rubber band / marquee selection:
All visualizations now support marquee selection as well as multiple selections
through Shift-Click operations.
- Save images of visualizations more
easily: Menu items have been added to each visualization that makes it
easier to save a PNG image of any SEURAT visualization.
- Can map some database columns to null:
When configuring SEURAT to access a 3rd party database if
there is no equivalent in that database for a concept SEURAT is able to
retrieve and analyze, then that concept can be mapped in the
pri_mappings.properties file (or equivalent) to always retrieve “null”.
- New structures only checkbox during
load into database: SEURAT can now be instructed, during load of data
from an SDF file into the SEURAT native database, to load only structures
and their identifiers. Previously the user had to make several changes
either to the SDF file to remove properties or in the file import to
database dialog to uncheck for import many columns.
Bug Fixes in SEURAT 4.2.2
- Corrected import of compound from
smiles for structure search; When an existing compound is loaded into
the compound sketching area to be used as a starting point for a structure
search and the compound comes from a smiles string, SEURAT now renders the
compound correctly by applying a clean2D operation.
- Added database columns sort correctly;
The data type of columns being added to an existing Assay Display from the
list of “addable” database columns is now controlled correctly by the type
designation given for that column in the pri_mappings.properties file (or
other mapping file for other databases). This means that sorting on
columns denoted as number will work as expected rather than sorting alphabetically
as they had done in the past.
- The axes of visualizations launched from
the primary Celera Visualization suite window now scale dynamically in
conjunction with dynamic filtering operations: Whenever a
visualization is plotting an axis that is also being filtered dynamically
the axis itself will change to plot only the remaining range based on the
filtering operation. Previously only the first scatter plot opened would
scale in this fashion.
SEURAT Version 4.2.1 March 9th 2007
Features added and improved in SEURAT 4.2.1
- Multiple Images Per Assay Display Cell;
SEURAT can now support multiple images per cell in the assay display when
configured to interpret a particular assays values as an image (either
through a direct URL or a call to a cgi script) and that assay has values
for multiple lots of a compound or multiple experiments against the same
lot.
- Obsolete and Activate Scaffolds;
Scaffold definitions can now be obsoleted / re-activated through menu
options in the File menu of the Assay Display
Bug Fixes in SEURAT 4.2.1
- Manual Scaffold Definition Re-enableld;
Several bug fixes and database schema changes were made to correct
problems reported with manual definition of a scaffold for use in later
analysis and visualizations. In Version 4.2 we had disabled the ability to
manually define scaffolds that appear in the Add Data Dialog but now this
feature has been enabled again.
- Test and Prod Server Same Machine;
The server can now be run multiple times on the same machine by providing
a different port number using the –Dserver.url=
//localhost:5247/CompChemServer and –Duse.server.url.only=true directives
during startup of the Server to specify a different port for the test
versus production servers. Previously due to a bug the server was ignoring
the –Dserver.url directive and using the same port number 5247 in all
cases.
- Test and Prod Client Same Machine;
The client can now be run multiple times on the same machine by providing
at least one or both instances of the client with the property –Dserver.url=
//localhost:5247/CompChemServer as well as the property
–Duse.server.url.only=true. Previously the SEURAT Client was only able to
connect using a per user set of preferences stored in the Registry on
Windows or a hidden directory on Unix meaning you had to switch connection
parameters between production and test each time you ran the client.
- Symbolic Link to Crystal Structure
Files Allowed; Symbolic or hard links on UNIX or shortcuts in Windows
can now be placed in the SEURATServer/sandbox/xtal directory in cases
where you want more than one compound to provide a link out to the same
set of crystal structure files. Previously links had been ignored.
- Correction to RS3 Structure Search
Script; rs3structurequery.pl script used to strucrture search RS3
databases now correctly returns the Compound ID rather than the Corporate
ID of matching compounds in a structure search.
- Forward Compatability of Load to DB
Profiles; Saved profiles for setting during a load of an SD or CSV
file into the database will now be forward compatible between different
versions of SEURAT.
SEURAT Version 4.2 February 7th 2007
Features added and improved in SEURAT 4.2
- SEURAT
can now be configured through multiple server configuration files, one per
database to read information from multiple databases simultaneously.
- Map SEURAT
to any medicinal chemistry database be it home grown or third party
(IDBS, MDL, CambridgeSoft, Accelrys)
- If
your assay data is split across different tables for different assay
classes (screening versus inhibition for example) simply provide N
configuration files and link them in a primary/sub database relationship.
The separate assay tables can be in the same or different databases.
- Load
partner, CRO or third party database data (e.g. ACD, WDI) from SD and CSV
files through the SEURAT Client into the Synaptic Science native database
and retrieve that data alongside your own corporate compound data.
- The SEURAT
Assay Display can now display images for Assay Values. This is done by
configuring SEURAT to interpret a set of assays designated by the SEURAT
Administrator as representing a URL that is then used to depict images in
the cells of the column for those assays.
- This
can be done either by having the assay value itself be a URL that will
produce an image or it can be done by specifying a base URL (typically a
CGI script) to which the assay value will be passed as the first and only
parameter
- The
size of the image so produced can be controlled by the same configuration
section in the SEURAT database mapping configuration file
- Data
can now be loaded into the Synaptic Science native schema in two ways.
- First
it can be loaded as was possible in this and all prior versions by
presenting the data in an SD or CSV file whereby each record (SDF) or
line (CSV) represented the Assay, Lot and Property data for a single
compound. We refer to this as pivoted data loading because the data has
been pivoted around the corporate identifier. Note that pivoted data loading corresponds to presenting data for
loading in a format almost identical to that which is displayed in the SEURAT
Assay Display
- Newly
added in this version is the ability to load un-pivoted data. In this
mode an SD file is used to load only the structure information about the
compound. A CSV (or number of CSV) file is then used to load the Assay,
Lot and Protocol data pertinent to each experiment run against a
compound. This mode of data load allows a much finer grained control over
the lot and protocol information against which each assay result it
loaded. Note that un- pivoted data
loading corresponds to presenting data for loading in a format almost
identical to that which is displayed in the SEURAT Result Display
- Hide
rows can now be made to persist during a save job / re-run job cycle.
Previously a conscious decision had been made to not allow this capability
because it was thought that people would tend to forget they had hidden
data and think that SEURAT was producing incorrect results. Now:
- A
new configuration file property syn_allow_hide_row_persistence in the SEURAT
Server configuration file can be used to enable (set to true) or disable
(set to false) the persisting of hidden rows
- A
default height for the column headers in the Assay Display can be provided
in the SEURAT Server configuration file via the syn_default_column_header_height
property.
- If a
user subsequently alters the value and saves the job then the user
selected value for the height of the column header will override the
default for that job.
- SEURAT
has been upgraded to the latest version of JChem Base (3.2.3) from
ChemAxon bringing SEURAT up to date with the latest features and bug fixes
contained therein. In particular this allows the interactive loading of
compound structures through JChem Manager (also known as jcman) into the
syn_structure with designated identifiers for each structure. This means:
- Customers
who want to structure search using JChem Base but don’t want to load
their assay data into the Synaptic Science native database can now do so
and have SEURAT link out to their existing tables for assay and property
data in all other cases
- Batch
loading of customer data will not be possible until ChemAxon fix a bug
that is causing the linking of fields in an SD or CSV file to the cd_id
column of a JChem Base structure table from working. ChemAxon have
committed to providing such a fix in their next bug fix release.
- Don’t
forget to check out the property calculation capabilities accessible by
double clicking on a compound in the Assay Display and then selecting the
calculation of interest from the Tools menu of the resulting window.
- You
can run each such application once per SEURAT session. Additional
licenses are required to remove this restriction
- To
make the task of SEURAT administrators easier the database connection
parameters that formerly had to be provided by every client user can now
be configured in the SEURAT Server configuration file meaning:
- The
first time a user logs in they will automatically use those settings for
authentication and subsequent queries issued from SEURAT
- If
the settings need to change at any time the administrator can make the
change once in a centralized place and then each client need only click
on the “Set to Defaults” button of the “Prefs…” dialog of the client to
pick them up.
- The database
tabs that appear in the “Prefs…” dialog are driven entirely by the
contents of the xxx_mappings.properties SEURAT Server configuration
files.
i.
This makes what the user sees more
meaningful
ii.
The old tabs for “smpc”, “asc” and “tsi”
were removed as they have been replaced by properties within the SEURAT Server
configuration file(s) that indicate which database hosts each concept (smpc =
simple molecule property calculator, asc = available screening compounds, both
are tools that can be launched within SEURAT).
iii.
The tab names can now be made meaningful
to the user community.
iv.
NOTE: You must have a primary file named
pri_mappings.properties that contains the is_auth_db=true for backward
compatibility if you already have SEURAT jobs saved.
- A new
property syn_dont_average_assay_rules in the SEURAT Server configuration
files has been introduced to allow customers to specify which assay
results should not be average when multiple experimental results exist for
a given compound across lots. Most typically this will be a specification
of rules that identify Pharmacokinetics assay names. The rules allow:
- Designation
of startsWith, indexOf or endsWith logic for matching
- A
list of strings to which the startsWith, indexOf and endsWith logic will
be applied.
- The
Synaptic Science native schema has been upgraded to more widely track
which data comes from which file during the load of SF and CSV data and
has been improved in many other respects. If you use the Synaptic Science native database YOU MUST run the
appropriate update script for your PostgreSQL or Oracle database to be
able to use version 4.2 of SEURAT.
- For
Oracle the update script can be found in the SEURATServer/oracle
directory
- For
PostgreSQL the update script can be found in SEURATServer/postgresql or
can be launched form the Start->All Programs->SEURAT
Server->Update Database Schema menu item
- You
DO NOT have to run either update script if you are interfacing to your
own or a third party vendors database even though you include the minimal
set of SEURAT tables as no changes were made to the tables that comprise
the minimal set.
- More
user roles have been added to allow finer grained authorization control.
The new roles are:
- CanLoadToDB:
Can a user load SD and CSV data into a Synaptic Science native database
schema
- CanSeeOthersJobs:
Can a user open the “Other Users” folder, see other peoples jobs, run
them and then save a copy of the job of their own to work with.
- CanSeePKReports:
Can a user open and view PK reports when they are available for a
compound
- CanSeeXtal:
Can a user see and open crystal structures into PyMol for viewing.
- Added
a dialog to inform the use if the SEURAT Client cannot locate the SEURAT
Server upon startup that gives the user the option to alter their server
connection parameters
For more information about the particulars of how to set any
of the new features that rely on property settings in the SEURAT Server
configuration file please refer to the SEURAT Configuration Guide on our
website.
Bug Fixes in SEURAT 4.2
- Freeze
pane now works for all column types. Assay columns would not present a
right mouse click menu which includes the freeze pane option when the
underlying column data type was numeric. This has been corrected.
- The
Available Screening Compounds application now correctly connects to the
database specified by the SEURAT Server configuration file containing the
property setting “contains_asc=true”. Previously the ASC was trying to
perform structure searches against a database that was assumed to be on
the same machine as the client.
- During
data load from a file SEURAT used to incorrectly create a new Protocol
entry in the database for a given Assay for every result stored against
that assay. This has been fixed by the introduction of the ability to load
un-pivoted data into SEURAT from SF or CSV files.
- Un-pivoted
data is data in the same or very similar format to that which is
displayed in the Result Display of SEURAT where each experiment details
appears in a row and a compound can appear many times in many different
rows.
- Commented
out an erroneous reference to a non existent SVL file from within the
caclprops.py script that SEURAT uses to interface to property calculators
that do not have a Java binding. This issue only related to the
calculation of properties using the Molecular Operating Environment (MOE).
- Code
that was trying to speed up queries whereby the same assay or assays had
been added more than once as columns to a job by running only one of those
duplicates were not distinguishing between different types for the same
assay. Previously the addition of a second column for a different result
type for an assay would not work properly when the SEURAT job was
re-executed.
- Quick
Searches that show aggregate results (List all Projects, List All Assay
Result Types) that don’t require transformation of the Result Display into
the Assay Display would try to display the Assay Display anyway after a
save re-execute cycle. This made it appear as though not results had been
returned because the results were actually in the Result Display which was
closed to produce an empty Assay Display. The fix means only the Result
Display is shown.
SEURAT Version 4.1.2 December 19th 2006
Bug Fixes in SEURAT 4.1.2
- The
URL generated during the emailing of a link to a SEURAT job was mistakenly
being set to “localhost” rather than to the name of the machine upon which
a Web Server was running to serve up the SEURAT Client. The SEURAT
administrator can now specify the URL that will be used by SEURAT as the
base for linking to existing SEURAT jobs via an email message in a SEURAT
Server configuration file.
- The
Web Start client distribution mechanism was built atop of version 5 of the
jetty web server. In order to use this same jetty web server as the
mechanism for the support of the emailing of URL links to SEURAT jobs we
had to upgrade to the use of jetty version 6.0.2 or later. Users that wish
to make use of the jetty Web Server to enable the emailing of links to SEURAT
job’s need to upgrade to this new version of the Web Start Client by
downloading it from our website.
- Customers
that were using a compound identification scheme that, after the prefix,
contained non numeric characters were not able to import compounds into
the compound sketch area from within the Template Form or the Sigma tool
for column addition. This issue has been corrected. An example of a
compound that would not have worked prior to this bug fix is ABC-12345A
where the final A was the character that confused SEURAT.
- The
Quick Search query for notebook information was fixed to no longer always
reference the synaptic science schema column for the alternate_id for the
notebook and instead properly maps to whichever column is specified by the
customer in the syn_mappings.properties file.
- The
right mouse popup menu when over a cell in the Assay Display for a column
containing numeric assay values was not appearing. This has been corrected
such that all popup menu options for things like freezing a column into a
separate pane can now be performed on any column.
- The SEURAT
Client would, very infrequently and in a difficult to reproduce manner,
become unresponsive after the execution of an existing job from within the
List Manager or just after adding a new column to the Assay Display. Both
of these issues have been resolved.
Features added and improved in SEURAT 4.1.2
- Searches
including one or more compound identifiers made in the Template Form and Quick
Search dialogs can now include the “*” character to designate that a
wildcard match should be performed.
- The
sigma tool in the Assay Display now supports a new function called regex
which allows expressions that include Java regular expressions to be
evaluated in order to create a new column. For example, it would be
possible to parse just the year out of a date based column with the column
identifier of A that had values like 2006-05-12 with the following sigma
tool expression:
- Year
:= regex(A, "(\\d{4})-(\\d{2})-(\\d{2}).+", 1)
- More
log messages added to the Server log on particular client activities in
order to allow better auditing of which functions are being used within
SEURAT.
- Removed
implicit rule that placed PK assays into folder based on the portion of
the name after the last underscore into a folder with the name of that
last portion or a folder named “Other” if the PK assay had no underscore
in its name.
- Now
new assay folder rule entries are allowable that can designate an “ends
with” condition rather than a “starts with” condition so that all rules
pertaining to how assays are assigned to folders can be controlled from
within the syn_mappings.properties file
SEURAT Version 4.1.1 November 16th 2006
Bug Fixes in SEURAT 4.1.1
- Scientist
name when supplied during a load of SD data into the database was being
ignored and the name of the currently logged in user was being used for
the scientist that created each lot instead. This has been fixed so that
the currently logged in users name will only be used in the event that a
scientist property is not provided in the SD file being loaded.
- SEURAT
no longer produces a mixture of Oracle and ANSI outer join syntax when
communicating with Oracle. The ANSI outer join syntax is only required for
PostgreSQL and MySQL and in fact using a mixture of Oracle and ANSI outer
join syntax when communicating with Oracle causes problems.
- A new
assay column can be added for results in an Assay Display with > 1000
rows. A bug that prevented this was inadvertently introduced during the
changes to make SEURAT database agnostic.
- Externalized
remaining Synaptic Science native database schema specific Quick Searches
that were missed in version 4.1. Those were
- All
data on a notebook
- List
of all projects, their description and comment information along with the
count of assay results per experiment
- List
of all result types along with a count of how many results exist for each
assay type.
- When
invoking an external script or executable to perform a structure search on
a Unix platform SEURAT no longer incorrectly inserts “cmd.exe /c” before
the program being called which was behavior specific for executing scripts
on Windows. SEURAT now correctly switches on the type of operating system
and only includes the “cmd.exe /c” if the SEURAT Server is running on
windows.
- It is
no longer necessary to provide a seurat.ini override file for property
setting to the server when SEURAT is interfacing with an Oracle database
in order for the saving of new jobs to work. SEURAT now correctly reads
the database connection settings from Java preferences rather than from
the chemwb.ini file used to bootstrap the database connection properties
- Multiple
selection of compounds in the Human Kinome Tree Viewer although rendered
correctly as the activity of each Kinase on the Kinome Tree simultaneously
used to leave the viewer in an error state (unresponsive) until a single
selection was once again made. This issue has been resolved so that the
viewer will remain responsive during multiple compound selections.
- Corrected
several situations where combinations of settings in each of the tabs
within the Template Form (Assay & Compound, Lot
and Structure) would produce incorrect search results when one or more of
the tabs specified a search that should have caused the whole search to
return no results.
- This
was especially true of searches involving the structure tab in
combination with the other Template Form tabs when the other tabs had
settings that should have resulted in no results being returned.
- This
was a difficult bug to find because if the constraints in each tab
produced at least one row then the searching would work perfectly.
- Corrected
several situations where a job would return different results after it had
been saved and then was re-executed.
- Most
typically this would happen for a job that returned no results when run
initially but after being saved and then re-executed it would show some
results that were from the substructure search part of the query.
- Copy
of a SEURAT job’s constraints when
that job includes a property calculated on the client no longer throws an
Exception
- A
URL’s menu now correctly appears on a right mouse click over a cell in the
Assay Display that contains a valid URL or URL’s. Each URL is presented as
a separate menu item of the URL’s menu in the popup menu.
- Jobs
created from the Template Form that have corporate rows added no longer
return no results when saved and then re-executed in a subsequent SEURAT
session. NOTE: That a bug still remains that is stopping “add rows” from
working properly with jobs created by the Quick Search tool.
Features added and improved in SEURAT 4.1.1
- It is
now possible to set the assay type, assay units and concentration units en
masse not only to one of the values in the preconfigured list of values
but also to an arbitrary value entered by the user.
- Scripts
have been provided that can be modified by customers (to provide correct
paths to 3rd part executables) in order to enable the following
property calculations on a list of compounds from within the Assay Display
(note the need for MOE, ACDLABS and Strand Life Sciences batch software
licenses for these property calculations to work):
- In
addition a new section in the syn_mappings.properties file was added to
allow control of the display name for each of these properties and to
allow extension of SEURAT to support additional properties for which a
customer may already have their own property calculators.
- Customers
can now expand the list of Kinase name mappings known to SEURAT for use
with their own company proprietary, UPSTATE or AMBIT assays that are used
by the Human Kinome Tree Viewer to visualize Kinase selectivity.
- Customers
can also control the assay name and assay type used in the search for
compounds that have UPSTATE or AMBIT results
- The
Expanded view which shows assay results per lot aligned so that it is easy
to tell which lots for a given compound have been assayed has been
re-enabled. It had been disabled in the prior release (4.1) because of the
change of the lot id from a number to a string.
SEURAT Version 4.1 October 2006
NOTE: Due to incompatibilities in the way Lot
number is stored between version 4.1 (String) and earlier versions (Integer) it
is necessary to refresh any cached job results after installation of version
4.1
To do this simply select the Interactive Cache Builder
option AFTER having started the SEURAT server. This will cause any cached jobs
to be re-executed so that their format is compatible with version 4.1
Bug Fixes in SEURAT 4.1
- Add
Rows functionality now correctly retrieves the compound structure when a
corporate compound is added to a job that contains the results of loading
the contents of an SD or CSV file into a job.
- CSV
file load into the database no longer requires that the identifier column
be named “Corporate ID” and be the first column in the file. The selection
of the identifier column can now properly be made using the combo box at
the bottom of the file import dialog.
- File
import into database now does a much better job at parsing out standard
deviation and concentration information that is specified within the value
for an assay. Previously SEURAT only dealt well with concentration
information when it was supplied as part of the assay name. SEURAT was
failing to parse standard deviation information out of assay values.
- As
part of this resolution a decision was made to only parse the first line of
a multi-line property value from a property in an SDF file placing the
remaining lines content in the comment for the assay result. The
ramification of this is that SDF file properties that are numeric should
only contain a single value. If you want to provide multiple values for
different lots then the compound needs to be repeated in the file N times
with each repeat specifying different assay values and optionally the lot
number.
- Sorting
on columns will now work even if after a file load into the database a
mixture of number and string assay result values are created.
Features added and improved in SEURAT 4.1
- SEURAT
is now meta-data configurable. All database table and column references
have been externalized so that it is now possible to reconfigure SEURAT
through changes to the meta-data properties file (syn_mappings.properties)
to read from a database other than the native SEURAT database shipped with
the platform. See the SEURATPersistenceLayer.doc guide for details on how
this can be done.
- One
caveat is that file load into the database currently only works against
the native SEURAT database shipped with the platform
- It is
now possible to save your settings for a file import into the database so
that for file imports that occur often you do not have to reset the Assay
name, type, unit and concentration information each time.
- It is
now possible to set the assay type, assay units, assay concentration and
assay concentration units en masse by selecting several rows in the import
to database dialog and then using the popup menu from the right mouse and
selecting the appropriate menu item.
- SEURAT
now ships with scripts that can create the native SEURAT database either
in PostgreSQL 8.1 or Oracle XE. SEURAT can switch at runtime via
preferences between interfacing with a PostgreSQL or Oracle XE database.
- SEURAT
can now be configured to perform structure searches using a script called
by the SEURAT server rather than using the built in JChem Base solution.
This allows customers with their own structure search solution to continue
to use that solution within SEURAT.
- By
popular demand the type of the lot identifier has been changed from
integer to string allowing a lot more flexibility in terms of lot
identifier format. As a result of this change the Expanded View is
temporarily unavailable until the next minor release.
- The
rules for mapping assay names into folders can now be configured through
the syn_mappings.properties file. The default rules SEURAT has always used
are in the file as a starting point. Once changes have been made to the
way folders are mapped the SEURAT server will need to be restarted.
SEURAT Version 4.0.2 August 2006
Bug Fixes
- When
the test cMet dataset is loaded but the corporate prefix is set to other
than CRA then you could not:
a) Double click on a CRA id in the corporate id column to see all assay
results for that compound from within the Assay Display did not return any
results
b) Link to a crystal structure for a CRA compound that had a crystal
structure did not retrieve the crystal structure.
c) Quick search to "List all compounds with crystal structures"
did not return valid CRA compound ids.
- File
upload into the database could not be called from a client running on a
remote machine.
- Could
not initiate a sub-structure search from a remote client.
- Corporate
identifies were always given a hyphen character during install even if one
was not provided.
- Corporate
identifiers were always assumed to be zero padded to a fix length of
digits during server installation.
- Corporate
identifier and hypothetical identifier prefix could not be left blank.
- Complex
assay names produce an "Invalid Assay Format" exception when
added via the add column function in the Assay Display.
- PDF
export of Assay Display or Result Display results placed a "Celera
Confidential" as the header of each page.
- XLS
export does not produce a format compatible with Excel 2002.
- Cannot
save SAR display to a CWB format. This issue has only been partially
corrected because not all facets of the SAR display are reconstituted from
CWB yet, like for example coloring.
- Overwhelming
amount of debug messages generated to the client log file when you load a
file into the database after having visualized some results.
- Visualization
snapshots would not reopen when a new column was added or an existing
column renamed in the Assay Display upon which the snapshot was saved.
- Could
not specify "No Prefix" in the structure sketched views when
trying to retrieve a compound as the basis for a substructure search or
SMPC hypothetical compound addition.
- Notebook
ID designated within an SDF file as property or a CSV file as a column was
incorrectly being used as the id rather than alternate_id in the notebook
database table. This meant non numeric characters could not be present in
the notebook identifier.
- ASC
tool was not launching correctly from a remote client.
Features added in SEURAT 4.0.2
1. A new search with the assay name, type and units
automatically filled can be launched from the Assay Display via a right mouse
click with the cursor over an Assay column by selecting the "New
Search..." menu item from the resulting popup menu.
2. A new search with the cutoff value for the assay also set
can be launched from a the Assay Display via a right mouse click with the
cursor over a cell of any Assay column by selecting the "New
Search..." menu item from the resulting popup menu.
3. A new preferences tab was introduced named asc that
allows the database connection parameters for the asc tool to be set by the
user.
4. A new preferences tab was introduced named “Excel 2002
Converter” that allows the hostname and port of the machine running the
optional Excel 2002 Converter server to be specified by the user.