Information To Share With Applied Biosystems Regarding The ABI 3500 During Visit NIST BUTLER ABI3500 Update SWGDAM July2011
User Manual: BUTLER 3500
Open the PDF directly: View PDF 
.
Page Count: 20
Status Update on
ABI 3500 Open Letter
John Butler
SWGDAM
July 13, 2011
Open Letter Background and Response
•Last fall we were asked by NIJ to investigate the ABI 3500 Genetic
Analyzer, which is the newest instrument from Applied Biosystems
used for forensic (and clinical) DNA sequencing and genotyping. NIJ
provided part of the funds used to purchase an instrument, which we
obtained in November 2010.
•The federal government (through NIJ funds from Congress) is being
requested by states to purchase the ABI 3500. We have conducted
some validation studies and explored cost analysis at NIJ’s request
(see presentation given December 7, 2010 here at NIST:
http://www.cstl.nist.gov/biotech/strbase/pub_pres/ForensicsNIST-
ABI3500.pdf).
•I gave a presentation at the FBI’s SWGDAM (Scientific Working
Group on DNA Analysis Methods) meeting in January 2011 on what
we have found with the ABI 3500. A number of people requested
that I follow up and something be written by myself or
SWGDAM to Applied Biosystems to express our concerns.
Background (cont)
•Jeff Sailus (ABI) visited NIST January 28 and provided helpful information
on technical questions
•I was busy finishing my book and so did not write the letter until early March
2011. After getting approval of SWGDAM chair, I sent the letter to
SWGDAM membership on March 14. I also sent the letter to about 900
forensic DNA scientists around the world invited their support. Support for
the letter has been overwhelming and 101 people agreed to sign on. (a
number of other people who found out about the letter after it had been sent
to ABI have also expressed support)
•I sent the open letter to Applied Biosystems on March 31.
•ABI sent out a FAQ sheet (completed March 30) to customers with their
responses.
•A letter addressed to me was sent April 4 from Lenny Klevan (ABI HID
president). It has also been provided to many of their customers.
Background (cont)
•I gave a presentation at the ENFSI meeting in Brussels on April 8
discussing what had happened to that point.
•Because I was traveling and out of the office much of the time since early
April, I was unable to connect with ABI until April 25 when I spoke with
Lenny Klevan, the president of Applied Biosystems by phone for 45
minutes. He expressed desire to work with NIST to improve communication
with the community and invited me to come to their headquarters in Foster
City, CA or for several ABI scientists to come here to NIST to talk to us
about the issues.
•Melissa Kotkin (ABI Field Application Specialist), who was already
scheduled to come, came April 26 to work with Erica Butts and Becky Hill on
our ABI 3500.
•Lisa Calandro and Lori Hennessey (ABI) visited NIST May 11, 2011
•The bottom line is that communication has been improved with the
forensic DNA community and that the open letter started this process.
Open Letter to Applied Biosystems
on Concerns with ABI 3500
•3/14/11 - emailed ~900 forensic DNA scientists
(SWGDAM, forens-dna, ENFSI, EDNAP) inviting
them to sign onto a letter that will be sent to
Applied Biosystems expressing concern with ABI
3500
•Very positive response with 101 who agreed to
sign the letter
•Letter was sent March 31 to the president of ABI
and scientists involved with the ABI 3500
•Community being notified of ABI’s response
Concerns Expressed in Open Letter
•RFID tags
•New .hid file structure requires new
software
•Short shelf life of reagents –would
like to see data for expiration times
Hopefully a change will result…
A desire for greater communication with the community
–the 3500 FAQ sheet is a good start but does not
directly address all of the concerns raised
Reagent Shelf Life Data
•What is the normal expiration data for buffer,
polymer, arrays, and kits?
•What data is this based on?
•They do not have any data (business
decision)
What was learned from the May 11 visit…
•RFID over-ride is possible (their R&D lab has instrument that can use
“expired” reagents) – they are “considering” making this option available
•New software is required for 3500 .hid or .fsa files due to new file
structure
•They do not have ANY data to support short shelf life of 3500 reagents
–A business decision to set hard stops to keep labs from having failures that lead
to ABI having to replace arrays
•ABI 31xx instruments have DEPRESSED signal (i.e., should have a
lower analytical threshold)
•Normalization is not well worked out by ABI or really understood
(although this has been a major selling point for the 3500)
•ABI was shocked that there were concerns with some of the feedback
Cost for the Forensic DNA Community to
Switch from ABI 3100s to 3500s
1. Instrument up-front cost
–Within the U.S. funding requests will likely come from federal grants
2. New software purchase
–Will likely be requested from federal grant funds (NIJ)
–new .hid file format will not work on current software (GMIDv3.2)
–3500 will not create .fsa files with 36cm arrays (HID applications)
3. Validation time & expense
– Relative fluorescent scales are completely different…
4. Operational cost
– ABI claims that the running costs are equivalent to 3130s…
NIST Calculated Cost per Sample for
ABI 3130xl vs. 3500 and 3500xl Reagents
$0.79
$1.11
$1.50
$0.96
$0.00
$0.20
$0.40
$0.60
$0.80
$1.00
$1.20
$1.40
$1.60
Cost Per Sample
3130
3130 (AB
Assumptions)
3500
3500xl
Running two plates per day (10 plates per week)
ABI 3500 Identifiler
D8S1179
D21S11
D7S820
CSF1PO
D3S1358
TH01
D13s317
D16S539
D2S1338
D19S433
vWA
TPOX
D18S51
AMEL
D5S818
FGA
0.00
0.03
0.06
0.09
0.12
100 150 200 250 300 350
Mean Allele Size, bp
SD Allele Size, bp
Ladder
Sample
Locus Average
s(Ladder): 0.049 bp
s(Sample): 0.042 bp
ABI 3130 Identifiler Plus
D8S1179
D21S11
D7S820
CSF1PO
D3S1358
TH01
D13s317
D16S539
D2S1338
D19S433
vWA
TPOX
D18S51
AMEL
D5S818
FGA
0.00
0.03
0.06
0.09
0.12
100 150 200 250 300 350
Mean Allele Size, bp
SD Allele Size, bp
Ladder
Sample
Locus Average
s(Ladder): 0.044 bp
s(Sample): 0.051 bp
ABI 3500 Identifiler Plus
D8S1179
D21S11
D7S820
CSF1PO
D3S1358
TH01
D13s317
D16S539
D2S1338
D19S433
vWA
TPOX
D18S51
AMEL
D5S818
FGA
0.00
0.03
0.06
0.09
0.12
100 150 200 250 300 350
Mean Allele Size, bp
SD Allele Size, bp
Ladder
Sample
Locus Average
s(Ladder): 0.044 bp
s(Sample): 0.053 bp
ABI 3130 Identifiler
D8S1179
D21S11
D7S820
CSF1PO
D3S1358
TH01
D13s317
D16S539
D2S1338
D19S433
vWA
TPOX
D18S51
AMEL
D5S818
FGA
0.00
0.03
0.06
0.09
0.12
100 150 200 250 300 350
Mean Allele Size, bp
SD Allele Size, bp
Ladder
Sample
Locus Average
s(Ladder): 0.049 bp
s(Sample): 0.054 bp
Precision Studies at NIST
No significant
difference between
the 3130xl and
3500 for precision
n=16 ladders
n=6 samples
s(Ladder): 0.049 bp
s(Sample): 0.054 bp s(Ladder): 0.044 bp
s(Sample): 0.051 bp
s(Ladder): 0.044 bp
s(Sample): 0.053 bp
s(Ladder): 0.049 bp
s(Sample): 0.042 bp
ABI 3500 (Identifiler) ABI 3500 (Identifiler Plus)
ABI 3130xl (Identifiler) ABI 3130xl (Identifiler Plus)
A Sampling of Feedback I Have Received…
•People did not just sign the letter but many have an
opinion about the issues or concern about ABI
customer support (I have received >100 emails –
often with some very strong thoughts)
• “I think that the AB3500 related issues most likely
represent the beginning of a sea of problems,
against which every independent lab must take arms. It
is not up to the manufacturer of a machine to decide
the basic procedures of a lab - it is up to the lab”
(4/29/11)
• “I greatly appreciate your advocacy on behalf of our
community. Hopefully we will be heard.” (4/1/11)
Do the responders own 3500 instruments?
• “We have the ABI3500 in our lab since October 2009
and use is for non-human biological trace analyses (both
fragment analysis and sequencing). We share the
concerns you mention in the letter.” (3/28/11)
• “I would like to sign the letter-We are currently in the
process of validating a database-dedicated 3500
instrument and agree with/support your suggestions.”
(3/18/11)
• “Thanks a lot for the initiative. Although we recently
bought two of the instruments and many licenses of
the ID-X software I fully support your letter. So please
add my name to the signatory list.” (3/18/11)
General Support for Open Letter and
Concern for Economic Issues
• “I strongly believe that your initiative is excellent
and actually we (i.e. the forensics community)
should have done this long time ago especially
in situations where we may have certain issues
being “monopolized” in the market. Once I
argued, during one of our ENFSI DNA WG
meetings, that if we are not careful about certain
things and while we are all trying to standardize
our procedures etc we may find this difficult to
accomplish because of non-scientific issues
such as economic etc.” (3/21/11)
A Recent Example of Feedback
Sent: Monday, May 02, 2011 10:13 AM
Hi John,
Thank-you for the update as well as your ongoing efforts to support our community
vis-à-vis Applied Biosystems instrumentation.
Our AB representative Dawn Waltman was here for a visit Monday. She followed
up regarding your letter and AB response and I advised their response had
“missed the mark”. As a result she has scheduled a teleconference with her
supervisor later this week. I will be echoing the same sentiments you have
included in your letter and your follow up conversations with Lenny Klevan.
Regards,
Ray
Ray Wickenheiser
Laboratory Director
Montgomery County Police Crime Laboratory
Response from Dr. Robin Cotton
(shared with her permission)
Sent: Saturday, April 30, 2011 10:39 AM
Dear John,
Thank you for the information and the inclusion of the letter from Dr. Klevan. It is
clear that Dr. Klevan does not consider the substantial time and expense which
will be required for each forensic laboratory for instrument and software
validation.
The other point which I feel is significant is the need for the additional software
purchase. Since he states that the new software is compatible with .fsa files, I think
the company should make a software exchange available at low cost for any
lab purchasing the 3500 instrument. Many commercially available software
companies make new versions available at reduced costs to individuals or groups
already running an earlier versions. Because of the increased number of
technical changes the 3500 presents, the validation data may be more
extensive than was required for previous instrument change-over and thus the
validation time and cost to each laboratory will also be increased.
(page 1 of 2)
Response from Dr. Robin Cotton
(shared with her permission)
It would also be relevant to ask Dr. Klevan to provide figures for the number of
current 3500 users without the inclusion of paternity testing laboratories which are
all commercial operations. While I am an advocate for private laboratories (both
forensic and paternity), these facilities have the option to raise prices and
accommodate the need for increased validation time and expense in other ways
that do not require federal or other government support.
Additionally, in the Biomedical Forensic Science Masters program here at BU, we
feel it is important to teach our students using current instrumentation and
techniques. Introduction of this new instrument will affect many forensic
science teaching institutions, both undergraduate and graduate, as well as
all current forensic DNA testing laboratories. These institutions have
significantly less access to NIJ funding for large equipment and software than the
operating forensic DNA laboratories. Thus the effect of changes reach into the
educational institutions as well.
Regards,
Robin W. Cotton, Ph.D.
Boston University (page 2 of 2)
ABI Released Their Developmental
Validation Study Last Friday
Going Forward…
•We like the data coming from the ABI 3500.
•How much further should we push on getting
“expired” reagent data, RFIDs turned off, etc.?
