Release Note
Agilent DNA Analytics 4.0.76
New for DNA Analytics 4.0.76

Users can create filters and templates in the workflow mode
In the 4.0 release, users were not able to create filters while in the workflow mode; they could
only do so in the interactive mode.

Module specific license support
In the 4.0 release, concurrent licenses were not properly supported. With this release, individual
modules can support the use of a server license or a text license. This allows users to have a
concurrent license for one application module, and a single-seat license for a different one.

Track visualization improvement
In 4.0, genomic tracks loaded in the viewer were placed on the same plane, making it difficult to
view multiple tracks if they overlapped. In 4.0.76, the genomic tracks are spaced apart from each
other in order to make them more distinguishable.

Gene name ON/OFF toggle
Users can now turn the display of gene names on/off in reports from the preferences dialog. This
makes it easier to view data, and keeps the view from becoming overly confusing.

Configurable Interval Margin
To define an aberrant interval, it is sometimes desirable to add a “margin” to the interval – this
helps to accommodate the fact that aberrant regions may extend beyond the first and last probe
of an aberration call.
In CGH 3.5 a default margin of 200 bp was added to all intervals, accompanied by an option to
apply the margin symmetrically or asymmetrically. CGH 3.5 aberrant intervals started from 200
bp before the start position of first probe in interval and ended 200 bp from start position of last
probe (i.e. typically 140 bp from end position of last probe).
In DNA 4.0.76 there is an option, configurable within the CGHAnalytics.properties file, to control
this behavior. Length of the margin to be added, if any; and the manner in which it is to be added
can both be configured. Default settings are same as CGH 3.5.

Design 014693 added to the default data.

Track description added to all the default tracks

2 tracks (hg18_CNV_Track and hg18_miRNA_Track) removed from the
default tracks shipped with application.
Issues from DNA Analytics 4.0, fixed in 4.0.76:

Expression design gets added to CGH node for UDF.
With the fix, expression designs get added to the expression node.

Methylation: clicking on gene view resizes the window
With the fix, clicking on a gene will not resize the window.

Graphical common aberration results different from common aberration
text report for ADM2 algorithm
With the fix, common aberration results are the same in the graphical and text versions for the
ADM2 Algorithm.

Intervals are not rendered in UI but reported in the interval summary report
and also rendered in the table view.
With the fix, intervals are rendered in the UI, in addition to the summary report and table view.

Display of overlapping annotations of tracks is like Gene View.
With the fix, overlapping annotations are easier to read.

Wrong no of probes reported in interval penetrance report.
With the fix, the correct number of probes is reported in the interval penetrance report.

Handling of “dye flipped” arrays with ADM-2.
Since ADM2 utilizes the log ratio and log ratio error information, the log ratio columns for ‘flipped”
arrays needs to be flipped. However, no such flip is necessary for log ratio error column due to
structure of error model calculation. In DNA Analytics 4.0, the flipping was inadvertently applied to
log ratio error column as well. Due to this, the log ratio error was not being taken into
consideration and the aberration results were calculated using ADM -1 error model (i.e. the
results were correct as per ADM -1 model but the additional refinements using log ratio error
present in FE files was not being taken into account) This has been rectified in DNA 4.0.76.
Issues from CGH Analytics 3.5, fixed in DNA Analytics 4.0:
1. User wants to see more than 3 genes if they are present in a interval
Description
In 3.5 only 3 genes were reported in the text reports though more than 3 genes
were present in an interval.
Fix/Workaround
In 4.0, all genes are reported in the reports.
Notes
None
2. Probe based report shows the probes that ought to be filtered out by a filter
Description
In 3.5, the probe based report shows some probes (with log ratios) that should
be filtered out by an aberration filter.
Fix/Workaround
Fixed
Notes
None
3. For CBS algorithm, no of probes for all intervals are shown as 1.
Description
In 3.5, if interval based aberration text summary is generated for CBS algorithm,
then for all intervals no of probes is shown as 1 though there are more probes in
an interval.
Fix/Workaround
Fixed
Notes
None
4. GeneView to TableView link gets unsynchronized after you combine intra-array replicates.
Description
GeneView to TableView link gets unsynchronized after you combine intra-array
replicates
Fix/Workaround
Fixed
Notes
None
5. Design should be visible easily in experiment properties window.
Description
Design information was not available in the experiment properties or design
properties.
Fix/Workaround
Properties dialog is provided on the right click of the design.
Notes
None
6. Need indicator when design file is successfully imported
Description
None
Fix/Workaround
Design import summary dialog is available after the successful import of design.
Notes
None
7. UDF file with less than 10 rows of data is not getting imported.
Description
None
Fix/Workaround
Fixed
Notes
None
8. Split output files content into sizes that Excel can display.
Description
None
Fix/Workaround
Files are spilt into multiple files if no of rows exceed 64000.
Notes
None
9. Probe based aberration summary report does not output gene name anymore.
Description
None
Fix/Workaround
Fixed
Notes
Gene name column is added to probe based report.
10. Users want to see Graphical Common Aberration kind of output for filtered data.
Description
In 3.5, Graphical common aberration cannot be applied on top of the filtered data.
Fix/Workaround
Fixed
Notes
All applied filters are now considered while generating Graphical Aberration
Summary.
11. Allow Correlation results window to be minimized.
Description
None
Fix/Workaround
Fixed
Notes
None
Known Issues in DNA 4.0
CGH Module
1. Not all Genome builds can be loaded:
Symptom
DNA Analytics does not support genome builds other than Human, Mouse, and
Rat. Users can try to load other genome builds of their choice; however there
may some difficulties in handling other genome build annotations. As an
example:
- C.elegans, using roman numerals (i.e.: I, II, III etc.) in naming chromosomes,
cannot be supported.
- D.Melanogaster , using chromosome labels such as 2L, 2R, 3L, 3R, cannot be
supported.
Fix/Workaround
None
Notes
None
2. Margins at the end of aberrant interval:
Symptom
In order to accommodate the fact that aberrant regions extend beyond the first
and last probe of an aberration call, DNA Analytics allows for margins that
extend into the regions before the next (non-aberrant) probe. Within the
application, the asymmetrical addition of a 200bp margin on either side of any
aberration call is still the default selection in establishing aberration call margins
as part of aberration detection.
Our aberration detection algorithm reports the first and last probe of any
aberrant interval. Optimally, margins should extend symmetrically from (“away
from”) these two probes. To illustrate this concept:
In an aberrant interval, let’s call the first and last probe the ith probe and jth
probe respectively.
The real boundaries of the aberrant interval may start anywhere right after the (i1)th probe and ends anywhere before (j+1)th probe.
One natural choice is to choose the midpoint between the (i-1)th probe and ith
probe as the start of the aberration and midpoint between jth and (j+1)th probe
as the end of the interval.
This approach, however, has problem that when there is a large genomic gap
preceding or following the interval, the interval stretches halfway into that gap.
To control this situation, DNA Analytics uses an Interval Margin to control how
far the interval is allowed to stretch. This is configurable from
CGHAnalytics.properties file (found in config folder of installed directory) by
setting the field INTERVAL_MARGIN to desired value. The default value is set
to 200 in DNA Analytics 4.0
However, while adding the margin, there is a choice of whether to do this
asymmetrically or symmetrically. The asymmetric means that the margin at the
end of interval gets added to start position of the last probe and symmetric mean
the margin gets added to the end position of last probe
In the current version of DNAAnalytics, by default, it is not done symmetrically.
What it means, only the start position of the probe is considered. The margin
gets added to start position of the first probe and start position of the last probe.
That is, for a INTERVAL_MARGIN =200, if the probe lengths are 60 and there
is a gap preceding or following the interval, the interval would stretch 200 bp
before the start position of first probe and 140 bp after the end position of last
probe.
The option INTERVAL_FROM_END_POSITION in CGHAnalytics.properties file
controls whether symmetric or asymmetric version be used. This option is set to
false by default and to get the symmetric version one has to set it to true.
However, if the INTERVAL_MARGIN = 0, for the single probe aberration call,
the start and stop positions of the interval are going to be the same. This will
affect some algorithms like Interval Based Penetrance for single probe
aberration calls.
Thus if INTERVAL_MARGIN =0 then INTERVAL_FROM_END_POSITION =
true has to be set in order to avoid the above issue
Fix/Workaround
Notes
The field INTERVAL_MARGIN in CGHAnalytics.properties file controls the
margin to be added. One can change it from its default value 200
INTERVAL_FROM_END_POSITION in the file CGHAnalytics.properties
controls how the margin would be added. Setting this to true (from default value
of false) would give the symmetric addition choice as described above
Caution:
The setting INTERVAL_FROM_END_POSITION = true is recommended if one
wants to set INTERVAL_MARGIN = 0.
CGH3.5 Settings correspond to INTERVAL_MARGIN = 200 and
INTERVAL_FROM_END_POSITION = false which is the default setting
3. Linux Installer does not work with kernel version greater than 2.2.5:
Symptom
DNAx installer does not get launched on Linux platform version 5.x. It is getting
installed on earlier version (which uses Kernel 2.2.* or lower).
Fix/Workaround
The ZeroG installer has LD_ASSUME_KERNEL. If you want to install a product
on linux whose installer was built by ZeroG, you have to do vim -b
theinstaller.bin and then change
export LD_ASSUME_KERNEL=2.2.5
to
#export LD_ASSUME_KERNEL=2.2.5
Alternatively, one can use the command
cat install.bin | sed "s/export LD_ASSUME_KERNEL/#xport
LD_ASSUME_KERNEL/" > new_install.bin
And finally execute the "new_install.bin"
Notes
Known issue of Install Anywhere.
4. ‘Could not start JVM’ during application launch:
Symptom
On some systems having less available memory, while launching application
‘Could not start JVM.’ error is displayed and application does not launch.
Fix/Workaround
For Windows/Linux installer, reduce the heap size in the ‘DNA Analytics
4.0.72.lax’ file located in the install directory of the application to appropriate
value according to RAM available.
For MAC installer, reduce the heap size in the ‘info.plist’ to appropriate value
according to RAM available.
Notes
Reducing the heap size can make the application slower and user can get out of
memory error while doing analysis on large data set.
5. Sample Addition by Drag and Drop Issue:
Symptom
User selects the arrays from the data node and tries to drag them into the
experiment. Sometimes it is not added to the experiment in the first attempt.
Fix/Workaround
This is a sporadic issue. User will be able to add the arrays in the second
attempt.
Notes
None
6. Gene View Distortion:
Symptom
The gene view gets distorted sometimes for experiments with high density
multiple arrays and scatter plot turned ON
Fix/Workaround
Resize the gene view a bit. Gene view will get restored to normal.
Notes
None
7. Unable to start next process for importing arrays in background.
Symptom
Sometimes after successful import of FE files in background, user is not allowed
to start another FE import process in background, though first process is
complete. An error message is displayed stating "Some arrays are still getting
imported. Cannot start another process."
Fix/Workaround
Close the application. Go to process folder in the application install directory and
delete the file 'ArraysImportingInBackground'. Restart the application and import
the FE files in background.
Notes
None
8. Progress bar stops moving while importing multiple FE files.
Symptom
If multiple FE files are getting imported at once, then progress bar stops moving
after some time, which gives an impression that application has frozen.
Fix/Workaround
Though progress bar stops moving still the process continues and data get
imported.
Notes
None
9. Scatter plot tool tip is not shown in stacked view
Symptom
If stacked view rendering is on in the chromosome and gene view, then scatter
plot tool tip is not displayed.
Fix/Workaround
Switch to overlaid view or use data table to view annotations.
Notes
None
10. Second attempt of search probes in eArray does not open the required page.
Symptom
If user searched the probes through ‘Search probes in eArray’ option in
DNAnalytics and browser of search result is still open then on 2 nd attempt new
browser is opened and ‘The page cannot be displayed.’ error is displayed.
Fix/Workaround
Logout from 1st browser session and close the 1st browser window.
Notes
None
ChIP Module
1. DNA Analytics 4.0 only renders scatter plots in log scale with a fixed range of
0.25-fold to 4-fold.
Symptom
1.
Ratios that are higher than 4-fold, which are very common in ChIP-chip
data, are rendered as 4-fold. During visual inspection of the data, this
inadvertently under-reports the shape of peaks of some true binding
events.
2.
The visualization of the noise level is exaggerated. The baseline for “good”
ChIP-chip experiments can often vary between 1 to 2 log units, which look
relatively clean in dynamic raw- (non-log) - scale, but noisy in fixed logscale.
Fix/Workaround
None
Notes
Note that this issue is entirely limited to visualization, and that the algorithmic
results are not affected. It will be addressed in future versions of the software.
Preloaded Data
These are the datasets that are preloaded into the DNA Analytics 4.0 application:
Type
AMADID
Genome Build
# Probes (Size)
FE Files
CH3
014791
Hg18
199399 (200K)
No
CGH
3.0
Design113
568592115
2101915_h
g17
014698
(Both hg17
and hg18)
Hg17
6044 (6K)
Hg17
Hg18
Hg17 = 99785
(100K)
Hg18 = 99805
(100K)
CGH
CGH
014950
018897
Hg18
Hg18
43016 (44K)
230499 (244K)
No
CGH
018898
Hg18
227599 (244K)
No
CGH
014693
Hg18
237834 (244K)
Exp
3.0
Hg17
6044 (6K)
Exp
Exp
ChIP
Design113
568922571
319439176
_hg17
014850
014868
014706
Hg18
Mm8
Hg18
42816 (44K)
42993 (44K)
237198 (244K)
No
No
ChIP
014707
Hg18
237194 (244K)
No
CGH
41 3.0 files
1. US22502705_251469814934_S01
_CGH-v4_95_Feb07_1_1
2. US22502705_251469814934_S01
_CGH-v4_95_Feb07_1_2
3. US22502705_251469814935_S01
_CGH-v4_95_Feb07_1_1
4. US22502705_251469814935_S01
_CGH-v4_95_Feb07_1_2
No
No
41 3.0 files
No
System Requirements

Windows XP or Server 2003 with Pentium 4 or later or Mac OS X 10.4.x with JVM 1.5 or
Redhat Linux 9

2GHz CPU or greater, dual core CPU recommended

4GB RAM minimum (need more when working with 244K or higher microarrays)

40GB Free Disk Space

1280x768 Display minimum
Installation
Part 1. Install the software (Windows XP)
1. Start the installation.
-
If you are installing from a CD, put the CD into the CD drive. The setup program will start
automatically.
If you are installing from another location, run the installation program on the computer
on which you want to install the program.
2. Read the Introduction screen and click Next.
3. Read the License Agreement, click I accept the terms of the License Agreement, then click
Next.
4. In the Choose Install Folder dialog box, click Next to accept the default location for installed files,
or click Choose for a different destination.
5. In the Choose Shortcut Folder dialog box, select where you want the shortcut for accessing the
application to be displayed, then click Next.
6. In the Pre-Installation Summary dialog box, review the list of installation parameters, then click
Install.
7. When the Install Complete screen appears, click Done.
Part 2. Change memory settings, if necessary
If you want to maximize the speed of processing, you can change the memory setting for the “heap size”
of several processes. The heap size is controlled by two flags, –Xms<size> and –Xmx<size>. JVM starts
with -Xms amount of memory and can grow to a maximum of -Xmx amount of memory. The 32-bit
machine JVM does not support over 1400MB.
To change memory settings for running DNA Analytics 4.0
1.
In Notepad, open the file “.../Program Files/Agilent/DNA Analytics 4.0/DNA Analytics 4.0.lax”.
2.
Find the line “lax.nl.java.option.additional=-Xms1024m -Xmx1400m -Dsun.java2d.noddraw=true”.
3.
Change the flags to your preferable memory setting. For example, if you have 2 GB RAM, change the
line to read “lax.nl.java.option.additional=-Xms1400m -Xmx1400m -Dsun.java2d.noddraw=true”.
Note: Make sure the letter m is present at end of size and there is no space between
the number and m.
To change memory settings for Workflow mode
1.
In Notepad, open the file “.../Program Files/Agilent/DNA Analytics 4.0/config/config_workflow.properties”.
2.
Find the property, HEAP_SIZE=-Xmx1200m.
3.
Change “1200” to your preferable memory setting. For example, if you have 2 GB RAM, change the
property to HEAP_SIZE=-Xmx1400m.
To change memory settings for the background process for importing FE data files
1.
In Notepad, open the file “.../Program Files/Agilent/DNA Analytics 4.0/config/config_FEImport.properties”.
2.
Find the property, HEAP_SIZE=-Xmx512m.
3. Change “512” to your preferable memory setting as you did for Workflow mode.