Reproducibility: Standardize antibodies used in research

[George McNamara]

*****
To join, leave or search the confocal microscopy listserv, go to:
http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
Post images on http://www.imgur.com and include the link in your posting.
*****

I encourage everyone on the listserv to read Bradbury et al's comment in
Nature, and to pass it on to your colleagues who use antibodies and
similar reagents.

full text is freely available at

http://www.nature.com/news/reproducibility-standardize-antibodies-used-in-research-1.16827

Summary:

*To save millions of dollars and dramatically improve reproducibility,
protein-binding reagents must be defined by their sequences and produced
as recombinant proteins, say Andrew Bradbury, Andreas Plückthun and 110
co-signatories.*



--



George McNamara, Ph.D.
Single Cells Analyst
L.J.N. Cooper Lab
University of Texas M.D. Anderson Cancer Center
Houston, TX 77054
Tattletales http://works.bepress.com/gmcnamara/42

[Martin Wessendorf-2]

*****
To join, leave or search the confocal microscopy listserv, go to:
http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
Post images on http://www.imgur.com and include the link in your posting.
*****

George et al.--

On 2/22/2015 11:31 PM, George McNamara wrote:

> I encourage everyone on the listserv to read Bradbury et al's comment
> in Nature, and to pass it on to your colleagues who use antibodies and
> similar reagents.
>
> full text is freely available at
>
> http://www.nature.com/news/reproducibility-standardize-antibodies-used-in-research-1.16827 
>
>
> Summary:
>
> *To save millions of dollars and dramatically improve reproducibility,
> protein-binding reagents must be defined by their sequences and
> produced as recombinant proteins, say Andrew Bradbury, Andreas
> Plückthun and 110 co-signatories.*

Interesting idea, as long as you aren't doing research in an area for
which the soon-to-be required recombinant antibodies aren't available.

To quote from the article, "If these steps are taken, scientists will
not want to use unsequenced binding reagents, and the absence of
sequencing information will lead to market disadvantages for vendors.
The uncharacterized, unsequenced research antibody will become
obsolete."  --This proposal could have the effect of forcing most all
biomedical research down a set of predetermined, prescribed channels.  
Their goal is worth, but this policy strikes me as likely to have some
unintended bad consequences.

Martin Wessendorf

--
Martin Wessendorf, Ph.D.                   office: [phone]
Assoc Prof, Dept Neuroscience                 lab: [phone]
University of Minnesota             Preferred FAX: [phone]
6-145 Jackson Hall, 321 Church St. SE    Dept Fax: [phone]
Minneapolis, MN  55455                    e-mail: [hidden email]

[Martin Wessendorf-2]

*****
To join, leave or search the confocal microscopy listserv, go to:
http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
Post images on http://www.imgur.com and include the link in your posting.
*****

George et al.--

On 2/22/2015 11:31 PM, George McNamara wrote:

> I encourage everyone on the listserv to read Bradbury et al's comment
> in Nature, and to pass it on to your colleagues who use antibodies and
> similar reagents.
>
> full text is freely available at
>
> http://www.nature.com/news/reproducibility-standardize-antibodies-used-in-research-1.16827 
>
>
> Summary:
>
> *To save millions of dollars and dramatically improve reproducibility,
> protein-binding reagents must be defined by their sequences and
> produced as recombinant proteins, say Andrew Bradbury, Andreas
> Plückthun and 110 co-signatories.*
>

Interesting idea, as long as you aren't doing research in an area for which the soon-to-be required recombinant antibodies aren't available.

To quote from the article, "If these steps are taken, scientists will not want to use unsequenced binding reagents, and the absence of sequencing information will lead to market disadvantages for vendors. The uncharacterized, unsequenced research antibody will become obsolete."  --Not all antibodies are generated by companies--many of the most important are generated by researchers themselves.  This proposal could have the effect of forcing most all
biomedical research down a set of predetermined, prescribed channels, with only limited room for development of new reagents.  Their goal is worthy, but this policy strikes me as likely to have some unintended bad consequences.

Martin Wessendorf


--
Martin Wessendorf, Ph.D.                   office: [phone]
Assoc Prof, Dept Neuroscience                 lab: [phone]
University of Minnesota             Preferred FAX: [phone]
6-145 Jackson Hall, 321 Church St. SE    Dept Fax: [phone]
Minneapolis, MN  55455                    e-mail: [hidden email]

[Feinstein, Timothy]

*****
To join, leave or search the confocal microscopy listserv, go to:
http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
Post images on http://www.imgur.com and include the link in your posting.
*****

I could imagine a Œsoft¹ introduction working fine.  These will be far
more expensive than standard antibodies, at least at first, but large
vendors could easily introduce high demand products like anti-smooth
muscle actin as an alternative reagent for people who want to spend more
for extra validation and reproducibility.  The value for diagnostic
medicine would be quite high.   Eventually the economy of scale would make
these accessible to research labs.  As just another new and useful
technology I think it has a lot of use, but I cannot imagine journals
would ever demand that researchers use nothing but.  It will never be
practical to submit every possible type of antibody to cloning and in
vitro synthesis.  

Best,


Tim

Timothy Feinstein, Ph.D. | Manager, Core for
Confocal Microscopy and Quantitative Imaging
333 Bostwick Ave., N.E., Grand Rapids, Michigan 49503
Phone: [phone] | Email: [hidden email]







On 2/23/15, 10:13 AM, "Martin Wessendorf" <[hidden email]> wrote:

>*****
>To join, leave or search the confocal microscopy listserv, go to:
>http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
>Post images on http://www.imgur.com and include the link in your posting.
>*****
>
>George et al.--
>
>On 2/22/2015 11:31 PM, George McNamara wrote:
>> I encourage everyone on the listserv to read Bradbury et al's comment
>> in Nature, and to pass it on to your colleagues who use antibodies and
>> similar reagents.
>>
>> full text is freely available at
>>
>>
>>http://www.nature.com/news/reproducibility-standardize-antibodies-used-in
>>-research-1.16827
>>
>>
>> Summary:
>>
>> *To save millions of dollars and dramatically improve reproducibility,
>> protein-binding reagents must be defined by their sequences and
>> produced as recombinant proteins, say Andrew Bradbury, Andreas
>> Plückthun and 110 co-signatories.*
>>
>
>Interesting idea, as long as you aren't doing research in an area for
>which the soon-to-be required recombinant antibodies aren't available.
>
>To quote from the article, "If these steps are taken, scientists will not
>want to use unsequenced binding reagents, and the absence of sequencing
>information will lead to market disadvantages for vendors. The
>uncharacterized, unsequenced research antibody will become obsolete."
>--Not all antibodies are generated by companies--many of the most
>important are generated by researchers themselves.  This proposal could
>have the effect of forcing most all
>biomedical research down a set of predetermined, prescribed channels,
>with only limited room for development of new reagents.  Their goal is
>worthy, but this policy strikes me as likely to have some unintended bad
>consequences.
>
>Martin Wessendorf
>
>
>--
>Martin Wessendorf, Ph.D.                   office: [phone]
>Assoc Prof, Dept Neuroscience                 lab: [phone]
>University of Minnesota             Preferred FAX: [phone]
>6-145 Jackson Hall, 321 Church St. SE    Dept Fax: [phone]
>Minneapolis, MN  55455                    e-mail: [hidden email]

[George McNamara]

*****
To join, leave or search the confocal microscopy listserv, go to:
http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
Post images on http://www.imgur.com and include the link in your posting.
*****

Hi Tim, Martin and listserv,

the second graphic "Money down the drain" in the article

http://www.nature.com/news/reproducibility-standardize-antibodies-used-in-research-1.16827

estimates that $350 million out of $700 million spent in the US annually
on protein binding reagents is wasted on 'bad' antibodies. There are
several (perhaps many) companies that can generate and deliver
substantial protein AND DNA sequences for recombinant antibodies or
alternatives for $3,500 or less. So, if all $350M in bad spending were
moved to 'g&d', this would enable 100,000 new antibodies or antibody
like molecules to be made and delivered every year (and that scale would
drive costs down even more and increase options). Depending on
need/value of each new antibody, this could be exclusively distributed
by a company, available through many companies, or available as a $65
expression plasmid from addgene.org (and addgene sells some multiwell
plates of plasmids for less than $65/plasmid).

If US researchers continue to waste money on bad reagents, they should
not be surprised if NIH stops funding them. Similar changes may happen
in other countries.

It is now straightforward to sequence paired immunoglobulin VH and VL
(first abstract) or TCR Valpha/Vbeta (from single cells - not the only
papers):

http://www.ncbi.nlm.nih.gov/pubmed/23334449

DeKosky BJ, Ippolito GC, Deschner RP, Lavinder JJ, Wine Y, Rawlings BM,
Varadarajan N, Giesecke C, Dörner T, Andrews SF, Wilson PC, Hunicke-Smith SP,
Willson CG, Ellington AD, Georgiou G. High-throughput sequencing of the paired
human immunoglobulin heavy and light chain repertoire. Nat Biotechnol. 2013
Feb;31(2):166-9. doi: 10.1038/nbt.2492. Epub 2013 Jan 20. PubMed PMID: 23334449;
PubMed Central PMCID: PMC3910347.

http://www.ncbi.nlm.nih.gov/pubmed/24952902

Han A, Glanville J, Hansmann L, Davis MM. Linking T-cell receptor sequence to
functional phenotype at the single-cell level. Nat Biotechnol. 2014
Jul;32(7):684-92. doi: 10.1038/nbt.2938. Epub 2014 Jun 22. PubMed PMID: 24952902.


As for antibodies or TCR's that will be used therapeutically -- for
Ebola virus victims, for example -- it makes sense to use recombinant
human proteins instead of "humanizing" mouse antibodies.

See also my and others comments at

http://www.nature.com/news/reproducibility-standardize-antibodies-used-in-research-1.16827

and the Baskin editorial and Hewitt article (both open access) at

http://jhc.sagepub.com/content/62/10.toc

Sincerely,

George


On 2/23/2015 12:06 PM, Feinstein, Timothy wrote:

> *****
> To join, leave or search the confocal microscopy listserv, go to:
> http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
> Post images on http://www.imgur.com and include the link in your posting.
> *****
>
> I could imagine a Œsoft¹ introduction working fine.  These will be far
> more expensive than standard antibodies, at least at first, but large
> vendors could easily introduce high demand products like anti-smooth
> muscle actin as an alternative reagent for people who want to spend more
> for extra validation and reproducibility.  The value for diagnostic
> medicine would be quite high.   Eventually the economy of scale would make
> these accessible to research labs.  As just another new and useful
> technology I think it has a lot of use, but I cannot imagine journals
> would ever demand that researchers use nothing but.  It will never be
> practical to submit every possible type of antibody to cloning and in
> vitro synthesis.
>
> Best,
>
>
> Tim
>
> Timothy Feinstein, Ph.D. | Manager, Core for
> Confocal Microscopy and Quantitative Imaging
> 333 Bostwick Ave., N.E., Grand Rapids, Michigan 49503
> Phone: [phone] | Email: [hidden email]
>
>
>
>
>
>
>
> On 2/23/15, 10:13 AM, "Martin Wessendorf"<[hidden email]>  wrote:
>
>    
>> *****
>> To join, leave or search the confocal microscopy listserv, go to:
>> http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
>> Post images on http://www.imgur.com and include the link in your posting.
>> *****
>>
>> George et al.--
>>
>> On 2/22/2015 11:31 PM, George McNamara wrote:
>>      
>>> I encourage everyone on the listserv to read Bradbury et al's comment
>>> in Nature, and to pass it on to your colleagues who use antibodies and
>>> similar reagents.
>>>
>>> full text is freely available at
>>>
>>>
>>> http://www.nature.com/news/reproducibility-standardize-antibodies-used-in
>>> -research-1.16827
>>>
>>>
>>> Summary:
>>>
>>> *To save millions of dollars and dramatically improve reproducibility,
>>> protein-binding reagents must be defined by their sequences and
>>> produced as recombinant proteins, say Andrew Bradbury, Andreas
>>> Plückthun and 110 co-signatories.*
>>>
>>>        
>> Interesting idea, as long as you aren't doing research in an area for
>> which the soon-to-be required recombinant antibodies aren't available.
>>
>> To quote from the article, "If these steps are taken, scientists will not
>> want to use unsequenced binding reagents, and the absence of sequencing
>> information will lead to market disadvantages for vendors. The
>> uncharacterized, unsequenced research antibody will become obsolete."
>> --Not all antibodies are generated by companies--many of the most
>> important are generated by researchers themselves.  This proposal could
>> have the effect of forcing most all
>> biomedical research down a set of predetermined, prescribed channels,
>> with only limited room for development of new reagents.  Their goal is
>> worthy, but this policy strikes me as likely to have some unintended bad
>> consequences.
>>
>> Martin Wessendorf
>>
>>
>> --
>> Martin Wessendorf, Ph.D.                   office: [phone]
>> Assoc Prof, Dept Neuroscience                 lab: [phone]
>> University of Minnesota             Preferred FAX: [phone]
>> 6-145 Jackson Hall, 321 Church St. SE    Dept Fax: [phone]
>> Minneapolis, MN  55455                    e-mail: [hidden email]
>>      
>    

--



George McNamara, Ph.D.
Single Cells Analyst
L.J.N. Cooper Lab
University of Texas M.D. Anderson Cancer Center
Houston, TX 77054
Tattletales http://works.bepress.com/gmcnamara/42

[Vitaly Boyko]

*****
To join, leave or search the confocal microscopy listserv, go to:
http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
Post images on http://www.imgur.com and include the link in your posting.
*****

Dear List,
Have you compared and/or tested performance of OCZ RevoDrive 350, OCZ Z drive 4500 and Intel's SSD P3700 PCIe SSDs when handling 100 GB+ incompressible data sets for transient data processing coupled with 2-3 single, non-raid SSDs for multi-threaded data caching? Any feedback will be greatly appreciated.
Thank you.
Vitaly



 

[George McNamara]

*****
To join, leave or search the confocal microscopy listserv, go to:
http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
Post images on http://www.imgur.com and include the link in your posting.
*****

Hi Vitaly,

all the OCZ RevoDrives, and many of the other OCZ SSD's (Vertex, if I
recall correctly, but maybe multiple SSD models from OCZ),  I bought at
UMiami experienced horrible deaths. Some not long after purchase. These
were mostly in real tower PC's not piece of junk Dell mini-towers (which
were replaced with real tower PCs on their deaths).

Yes, I realize you plan to use it for transient data processing. So
what? Buying junk from bad companies is not worth it.

Consider Intel P3600 or P3700 (or other model) "data center" quality
SSDs, or equivalent from other (non-OCZ) SSD manufacturers.

Also pay attention to the controller. I had a personal Adaptec SCSI RAID
controller have a horrible death years ago - no more Adaptec stuff after
that. 10(plus) years ago at CHLA I had our HP server (now quaint ~80 Gb
SCSI RAID) die a temporary death until HP replaced the dead SCSI
controller and simultaneously another component ... until they double
replaced, one defective component killed replacements of the other
components. After that server was functional we were able to send it off
to the CHLA data center for management - my office was quieter after
that moved out.

Data R.I.P.

best wishes,

George
p.s. I routinely backed up the UMiami imaging core on portable USB
drives (initially 0.5T, later 1T, 2T) ... had a shelf(s) full after 5
years.

On 2/24/2015 9:50 PM, Vitaly Boyko wrote:

> *****
> To join, leave or search the confocal microscopy listserv, go to:
> http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
> Post images on http://www.imgur.com and include the link in your posting.
> *****
>
> Dear List,
> Have you compared and/or tested performance of OCZ RevoDrive 350, OCZ Z drive 4500 and Intel's SSD P3700 PCIe SSDs when handling 100 GB+ incompressible data sets for transient data processing coupled with 2-3 single, non-raid SSDs for multi-threaded data caching? Any feedback will be greatly appreciated.
> Thank you.
> Vitaly
>
>
>
>
>
>    

--



George McNamara, Ph.D.
Single Cells Analyst
L.J.N. Cooper Lab
University of Texas M.D. Anderson Cancer Center
Houston, TX 77054
Tattletales http://works.bepress.com/gmcnamara/42

[Doube, Michael]

*****
To join, leave or search the confocal microscopy listserv, go to:
http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
Post images on http://www.imgur.com and include the link in your posting.
*****

Hi Vitaly,

Maybe this will be of interest:

"Tests show modern SSDs can handle a thousand years of use"

http://www.engadget.com/2014/12/05/ssd-2-petabytes-of-data/

I have a couple of Samsung 840 Pros in RAID 0 which can write at something like 800 MB/s - 1 GB/s. At least, fast enough to write a stream of USB3 data without choking. So far, so good. Also have OCZ Vertex 4s in a similar setup which have been fine so far - mostly I use these as a large temp cache for camera capture and tomographic reconstruction. I haven't done a serious benchmark comparing the two.

Michael

>> Dear List,
>> Have you compared and/or tested performance of OCZ RevoDrive 350, OCZ Z drive 4500 and Intel's SSD P3700 PCIe SSDs when handling 100 GB+ incompressible data sets for transient data processing coupled with 2-3 single, non-raid SSDs for multi-threaded data caching? Any feedback will be greatly appreciated.
>> Thank you.
>> Vitaly<http://www.rvc.ac.uk>

This message, together with any attachments, is intended for the stated addressee(s) only and may contain privileged or confidential information. Any views or opinions presented are solely those of the author and do not necessarily represent those of the Royal Veterinary College.

[Vitaly Boyko]

*****
To join, leave or search the confocal microscopy listserv, go to:
http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
Post images on http://www.imgur.com and include the link in your posting.
*****

Hi George,
yes, we had the same problem with the OCZ. The RevoDrive 350 has a cooling added, so it will not burn out as often as it would happen with the older RevoDrive models (3 X2, etc.). However, the main problem that I do not see any improvement in file Open/Save SSD versus 7200 rpm HDD when the file size exceeds 12 GB. Having 3-4 drives RAID 0 may improve it 2-3 fold. However, I have never compared, let say, 2-3 Raid Edition drives (Western Digital) versus TRIM enabled RAID 0 array of SSDs on a x8 PCIe card. Same applies to the Intel's DC P3700. However, $2,600.- for a 800 GB P3700 is too much and I have some doubts that it will outperform three Western Digital RE4 drives in RAID 0. I agree that I have to spend extra for a good x8 RAID controller (again in the range of ca. $1-2K). Thus, I am asking if someone has tested performance of PCIe x8 RAID 0 SSDs, NVMe SSDs in comparison to three Raid Edition HDD in RAID 0 configuration with a good RAID controller.
I hope to hear some feedback, as both Bitplane and Amira (that was "negligently swallowed" by FEI) have no funds and/or brains to test and suggest an optimal configuration for handling large data sets (20 GB- 2 TB). It is a bit strange that Light Sheet Microscopy people stay silent on the issue. Now both 3i and Zeiss have the Betzig's patent on Lattice Light Sheet. I understand that they (especially Zeiss) "pretend as if they know something BIG".
Best regards,
Vitaly   

     On Tuesday, February 24, 2015 11:34 PM, George McNamara <[hidden email]> wrote:
   

 *****
To join, leave or search the confocal microscopy listserv, go to:
http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
Post images on http://www.imgur.com and include the link in your posting.
*****

Hi Vitaly,

all the OCZ RevoDrives, and many of the other OCZ SSD's (Vertex, if I
recall correctly, but maybe multiple SSD models from OCZ),  I bought at
UMiami experienced horrible deaths. Some not long after purchase. These
were mostly in real tower PC's not piece of junk Dell mini-towers (which
were replaced with real tower PCs on their deaths).

Yes, I realize you plan to use it for transient data processing. So
what? Buying junk from bad companies is not worth it.

Consider Intel P3600 or P3700 (or other model) "data center" quality
SSDs, or equivalent from other (non-OCZ) SSD manufacturers.

Also pay attention to the controller. I had a personal Adaptec SCSI RAID
controller have a horrible death years ago - no more Adaptec stuff after
that. 10(plus) years ago at CHLA I had our HP server (now quaint ~80 Gb
SCSI RAID) die a temporary death until HP replaced the dead SCSI
controller and simultaneously another component ... until they double
replaced, one defective component killed replacements of the other
components. After that server was functional we were able to send it off
to the CHLA data center for management - my office was quieter after
that moved out.

Data R.I.P.

best wishes,

George
p.s. I routinely backed up the UMiami imaging core on portable USB
drives (initially 0.5T, later 1T, 2T) ... had a shelf(s) full after 5
years.

On 2/24/2015 9:50 PM, Vitaly Boyko wrote:

> *****
> To join, leave or search the confocal microscopy listserv, go to:
> http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
> Post images on http://www.imgur.com and include the link in your posting.
> *****
>
> Dear List,
> Have you compared and/or tested performance of OCZ RevoDrive 350, OCZ Z drive 4500 and Intel's SSD P3700 PCIe SSDs when handling 100 GB+ incompressible data sets for transient data processing coupled with 2-3 single, non-raid SSDs for multi-threaded data caching? Any feedback will be greatly appreciated.
> Thank you.
> Vitaly
>
>
>
>
>
>   

--



George McNamara, Ph.D.
Single Cells Analyst
L.J.N. Cooper Lab
University of Texas M.D. Anderson Cancer Center
Houston, TX 77054
Tattletales http://works.bepress.com/gmcnamara/42