ROBOT REPLY

Stuart Stuart
Fri Apr 25 10:43:39 EST 1997


ROBOT REPLY

Hello All,

At last, the reply to the question I posted all those months ago (!) 
summarised for you.  First up I'd like to thank all those people who 
replied to my postings it was good of you all to take the time and 
effort to write to me and volunteer your opinions.  Next up I suppose 
I'm going to have to include all the usual disclaimers (you know the 
story by now) - this document does not represent the opinion of my 
employers, some of it doesn't even represent my own opinions...I'm 
only passing them on and removing the names.  I've tried to make 
everything fair, but not everybody out there with robots replied, so 
some sections may slant one way or another a wee bit!  In the last few 
months I've spoken to people seen machines, and even tried one of two 
and the best advice I can give is that if you are looking for a system 
talk to your local reps and see for yourself.  In the mean time feel 
free to use this as a guide to what other people out there are using.

Oh, just before the main body of this message, in the course of this I 
spoke to someone who reminded me that what these machines actually do 
is mechanise the process of template prep, reaction set-up, and 
cycling (depending upon the platform used).  None fully 'automate' the 
process into one 'dynamic whole'.  Perhaps a solution to this would be 
to PCR up the fragment of interest, and then go into sequencing...it 
could all be done in a stand alone system...perhaps this should be 
kept in mind!

Here goes...(in no particular order)

Biomek 2000

There is a dye terminator (DT) protocol fully functional for the 
Biomek, and now DNA preps available too apparently you can perform 
plasmid isolation for $0.45 instead of $3-$4.  This platform is quite 
easy to use, but it should be noted that labware specs need to be 
fairly exact - tips can be bent.  Also, comes with a variety of 
'tools' - although the 2005L acts more like a 1255L tool when used 
with barrier tips.  Also there were some reports from users with 
respect that handling volumes below 55L does not fair well in some 
peoples hands.

The Biomek set-up time is roughly 25 minutes for 192 sequencing 
reactions with single primer from a master mix.  Perhaps useful with 
shotguns but set up time for anything else maybe too great (too many 
variables).

The Biomek is actually designed as a high throughput repetitive 
96-well based screening platform, so varying numbers of samples on a 
day to day basis may present a problem.  The machine itself is said to 
be reliable, and doesn't miss wells too often.

The operating software is OK - being the usual, point-and-click type 
which makes operation easier for first-timers!

Overall, a good, versatile platform, but some users point to it being 
perhaps too slow for anything other that repetitive 96 well shotgun 
work.


ABI 877

There are two types of robot in circulation from the producers of the 
373a and 377 (oh! and the 310...) - that Catalyst and the 877.    The 
catalyst (which I believe no longer available) was blessed with a not 
too great user interface which was very inflexible.   However reports 
allege that the 877 has much improved software, it now being an easy 
to use user interface (again of the point and click variety), the 
overhaul resulting in a more flexible platform than the Catalyst.

This platform has a high throughput capability, and seems to be 
primarily designed to be used for high throughput genotyping projects 
that require each sample to be run with multiple markers (either one 
at a time, or multiplex) and the software reflects this.  In general 
the software is liked by most people - being clear and easy to follow.

The 877 seems to have been the only machine anyone reported as being 
happy pipetting volumes below 35L with...as the 877 seems to have been 
specifically designed to be used as a small volume pipetter and, with 
ever increasing cost saving needed as well as better enzymes and dye 
incorporation, smaller volumes may become more vogue  than at present.   
It should also be noted that there is no disposable tip option with 
this platform - just a steel needle which, although ABI say no carry 
over is present, some people report worries over.  Also there is no 
multi-tip option unlike other platforms.

Other reported problems  seem to be mainly based around leaks and 
suchlike - although a fairly major problem with M13 forward primer 
reaction efficiency (and accompanying artefact peak at 70-80 bases) 
that are unsolved at present, but are possibly due to cycler block 
contamination.  At the time of writing the solution to this is 
unknown.


Packard Muitiprobe 204DT

Not too much feedback on this particular machine, but what did get to 
me indicated that it's a good platform for DT set up ( you can use 
fixed tip or disposable) but can only be used down to 35L with the 
small disposable tips, after this volumes are likely to be 'some' 
rather than a specified amount).  Liquid sensing helps when doing DNA 
preps, but not so much for DT set ups.  Possibly some stability 
problems, but is cheap.

Tecan Genesis

Again, not too much feedback for this platform either, again it is 
reported as being good for reaction set up.  As for the use of the 
Tecan it has a cumbersome9 user interface and programming interface.   
Pipetting below 55L could pose problems unless the tip touching 
function is used (then goes down to 25L).  Set up time is about 1hr 
for 96 reactions.  Very reliable and very flexible, but perhaps it 
might be quicker to do some things by hand!

Qiagen Biorobot

The 9600 is an automated template prep machine from Qiagen.  It brings 
with it the reputation of the company that produces one of the most 
popular DNA prep kits.  This robot is basically an mechanisation of 
the manual prep method, which uses the Qiagen kits on top of a vacuum 
manifold to speed things up.  

The 9600 works in multiples of eight or in a 96 plate format.  The 
biggest gripe about the machine seemed to be that in the set-up of 
samples the software was a bit cumbersome, as it isn't of the 'point 
and click' variety (you have to write protocols) which means that 
without extensive effort you can't run samples with different primers 
with ease.  However, all this is supposed to change with an upgrade in 
the software (due soon), along with the addition of a pump which will 
halve the current prep time of 96 samples.  The 9600 does require some 
hands-on time mid run - to change over the columns etc. etc.

The machine is also capable of setting up sequence reactions, and any 
other pipetting reactions. This system is also due an upgrade, with a 
disposable tip option being incorporated.



Autogen

The Autogen robot gives fairly good results but are slow in comparison 
with newer types of machine (and maybe more expensive too).  Problems 
occur in their inflexibility with regard to change in demand - 
protocol development is not easy as they are hard wired into the 
system.  NB This is a miniprep machine and does not set up PCR 
reactions.

Vistra Labstation

Again, this reply produced both positive and negative feedback.  This 
platform is this only fully integrated machine that will go from cell 
pellets to finished reaction without coming off the robot at any 
point.

The Vistra  uses a magnetic purification system (called FMP) to 
isolate the DNA (quick point from someone in the survey - if this is 
such a good isolation method is anyone using it manually??).  The  
purified DNA is then added to a sequence reaction, which is completed 
on the machine.

There are protocols available for both M13 and ds plasmid work, 
however the success of them seems to be an issue with some reports of 
the ds template prep giving problems (one site seems to do manual 
plasmid preps and use the labstation for it's ss isolation, then set 
up all reactions on the labstation, as the ds isolation gives them 
problems).  Another report says that it is quicker to perform DT 
reaction set-up by hand rather that use this robot.  On the other hand 
some people seemed happy with it, and use it for both ss and ds work.



More information about the Autoseq mailing list