There are a lot of things to consider before getting into proteomics.
Proteomics is not a turn key operation as many people would have you
believe. There are many, many dependent steps that need to be done almost
perfectly for a proteomic lab to be successful. Here are a few things to
consider, then I'll try to answer your questions.
1) What type of throughput will you be doing? This make a huge difference.
A few gels a week means your set-up can be relatively simple and you can
easily track samples, spots and mass spectra. Several gels per week begins
to become troublesome. Not because of running the gels, as that is the easy
part, but how to handle all the gels and keep track of samples, gels and
data. Simply analyzing gel images can take a large amount of time and it is
critical to have it done well and reproducibly. Also reproducibility
becomes very important in any effort above a few gels a week. If you intend
to run hundreds of gels per week, then you are competing with some very big
hitters in the proteomics world and nothing I can tell you would prepare you
for that endeavor.
2) Resources. It is very clear that doing proteomic analysis requires
several different scientific specialties. As I hinted above, a sample and
data tracking system will be extremely helpful if not mandatory, depending
on throughput. Bioinformatics capabilities will also be extremely helpful
if not mandatory. You will need the ability to search databases quickly
and, more than likely, automatically. Obviously you need someone to run
your gels. Gel quality and reproducibility are very important. You will
need scanning capabilities. Fluorescent imaging is the current state of the
art and these machines are a bit expensive. Many people still use Coomasie
and Silver staining. Analysis of the images will take lots of time.
Virtual gel skewing needs to be done correctly and this takes some real
"feel", especially if you are comparing more than 2-4 gels at a time.
Automated spot cutters and digestion robots are available, but can be
expensive. For brevity, I'll assume that you have your mass spec
capabilities in order and you are ready to analyze the samples that you
intend to process. From the above paragraph, the summary should be that
proteomics involves many different tasks and techniques. Anyone intending
to do proteomics should be prepared to either have themselves or be able to
bring in expertise in these different areas.
Now to your questions.
Gel systems. There are a few different gel systems to work with. I think
that the majority of people doing 2-D gels as the front end to proteomic
analysis are using either the Amersham/Pharmacia Multiphor or IGPhor systems
or the BioRad IEF system. I have used both the Multiphor and BioRad
systems. They are very similar and I think that the BioRad system is a bit
easier to run and runs slightly faster. I think that Amersham/Pharmacia has
the best IPG strips, but I use BioRads because they are very plainly labeled
and having mix-ups due to inserting strips backwards is minimized. For the
second dimension, use whatever you feel comfortable with. We use BioRad
mini and full sized gels and they work well for us. Reproducibility is the
key to running 2-D gels. Use a system that you can produce high quality
gels again and again and again. It is my strong opinion that the system is
much less important than the person running it. Pay absolutely fanatical
attention to detail and cleanliness to have the reproducibility you need and
remove any doubt of contamination. I can not stress that enough. Fanatical
attention to detail and cleanliness. Words for any Proteomer to live by.
Image capture. This all depends on your method of imaging. For visible
staining of gels, get a decent scanner type instrument. BioRad makes a good
one as do many other companies. There are very few differences between
different visible scanners. Fluorescent imagers are also available. I
personally like the Molecular Dynamics Typhoon system and have one in our
lab. BioRad's version is a bit clunky and not as versatile in my opinion.
I have not, in person, seen either a Kodak system or a Fuji system. Based
on what I have seen and heard from colleagues, I would view these 4 systems
as the top of the line manufacturers of fluorescent imagers.
Image analysis. This is a rapidly growing area with many products on the
market and rapid upgrades to software. That is a very good thing. I am a
big fan of Melanie and have a copy now that I hope to upgrade soon. Melanie
is very powerful and, in my opinion, does a better job of gel skewing than
others. I have used the Amersham/Pharmacia software (I forget the name)
and, as of 1.5 years ago, I was not that impressed. I have also used
RioRad's PDQuest and thought it was easy to use, but did not have many
features of Melanie nor the power of Melanie. I have used a program called
BioImage (I think it's from ESI?) and I was not impressed. It was not up to
the standards set by other programs in my opinion. They may have upgraded
it since I used it, I don't know.
Spot pickers. I do not have any recommendations for you, as I am currently
looking for one also. Make sure that whatever you get, it holds the gel in
one place and does not allow the gel to shrink or swell during the picking
process. Also make sure that the robot checks the gel for landmarks or
features so that the XY coordinates it uses correspond to the ones you
determined during the image analysis phase. This is critical and not very
easy to do.
Good luck and keep in mind that everything above are just my opinions and
every person/lab needs to figure out what works for them.
Jim Lawrence
Pioneer Hi-Bred Intl.
7300 NW 62nd Ave
Johnston, IA 50131
515-270-5909
> -----Original Message-----
> From: Molecular Resource [mailto:molecular@njc.org]
> Sent: Friday, June 30, 2000 4:25 AM
> To: Recipients of ABRF List
> Subject: PROTEOMICS
>
>
> WE ARE LOOKING FOR OPINIONS ON 2-D GEL SYSTEMS, ANALYSIS AND
> SPOT PICKING. WE ARE JUST BEGINING AND WOULD APPRECIATE ANY INPUT.
>
> THANKS, AMY MARRS
> NATIONAL JEWISH CENTER
> DENVER, CO
>
This archive was generated by hypermail 2b29 : Thu Jul 20 2000 - 09:37:01 EDT