Re: Genetic tests to do on foodstuffs to test PCR

[TRolandB]

Hi there, 

The problem is that sometimes the procedure works and we get the correct bands showing up in electrophoresis, while other times we get no bands at all. We always follow the same general procedure, sometimes varying the quantities of reagents and PCR times to try and get it to work. Repeating the exact procedure from the times it did work doesn't lead to it working again.

As the extraction process is the biggest variable we assume this is at fault. At the moment we extract it from cheek cells by scraping with a pipette tip. I've linked to the project page if anyone is interested, all info is there.

http://wiki.london.hackspace.org.uk/view/Sex_typing_with_amelogenin 

Cheers

Will

On Thursday, 28 June 2012 13:59:39 UTC+1, phDIY wrote:

Can you elaborate on your problems?

On Wednesday, June 27, 2012 6:12:10 PM UTC-5, TRolandB wrote:

Hi everyone, our group (London hackspace) have been doing some genetic tests (sex typing) recently, with PCR and electrophoresis. We're getting very inconsistent results, and think our extraction process is at fault, so we want to try some similar techniques but with DNA from foodstuffs, so we can be sure of getting the extraction right.

Does anyone have any suggestions of tests you can do that involve PCR and electrophoresis of this type?

Will

[Ethan]

Have you been running a ladder along with each gel or some other positive control? That would definitively tell you whether the error is in the extraction/PCR or the running/staining of the gels. What size is the fragment that you are trying to amplify? From my experiences, one hour at 100V seems to be a fairly long time, but that might be the result of differences in fragment size that I have worked with and that you are trying to look at.

As for tests with foodstuffs, I remember doing a PCR experiment in high school to test for the presence of a sequence that is commonly part of GMO crops. Unfortunately, I do not remember any of the specifics of what the sequence was.

[Cory Tobin]

The first step would be to really figure out which step is the
problem. Have you tried doing multiple PCRs on the same DNA
extraction?

My suggestion would be to do 3 different DNA extractions all from the
same person. Then do 3 different PCR reaction on each sample, mixing
the PCR reagents separately for each reaction, then (assuming you are
confident that your thermalcycler is consistent across wells) run all
the reactions at the same time and then load them all into the same
gel side by side. From there you should be able to tell if there is
variability between extractions, between PCRs, or both.

Also, what exact protocol are you using for the extraction? Your wiki
links to a couple of different protocols. If you're using
phenol/chloroform I would suggest doing multiple ethanol
precipitations as phenol inhibits PCR.


-cory

[TRolandB]

Thanks to all for the advice. In reply

- The ladder always works, so no problem with the electrophoresis I would think.

- The fragment sizes are 977 and 788bp,  quite large fragments I think?
- Good idea on the multiple PCRs, we may give that a go.

- on the page I link to there's a section detailing our extraction protocol: complete protocol as of June 2012. - we use chelex.

[Steve Frese]

1. That ladder suggests that you need to run the samples longer.  It's not separating effectively.  The gel size and the amplitude of the current also affect how much 'oomph' you need to pull the bands through the gel.  Consider extending this time, if possible.  This will also help you separate two sized bands in males (XY).

2.  yes to the other comment on etoh extractions.  phenol will kill your pcr.

3.  where are you getting the reagents for the PCR reaction?  if its a kit (or even ask for a free sample from some companies) follow their instructions if you aren't already.  I've found them to be more robust in their guidelines than old papers (i assume its old since it lists the dNTP concentrations, and I don't see that anymore in recent papers).  

For the Sigma kits that are linked on the wiki, these are hot-starts, and require something like a 5min denaturation to remove the antibody (or whatever it is in that kit) that prevents non-specific products from forming before the reaction 'should' begin.  it's a great feature, but if you don't account for that in your pcr reaction, you'll get ambiguous bands... and none of the gels really convince me you/they have a product yet (on that wiki).

if my suggestions dont make sense, just let me know.

--
You received this message because you are subscribed to the Google Groups "DIYbio" group.

To view this discussion on the web visit https://groups.google.com/d/msg/diybio/-/xkgSsplQ_l4J.

To post to this group, send email to diy...@googlegroups.com.
To unsubscribe from this group, send email to diybio+un...@googlegroups.com.
For more options, visit this group at http://groups.google.com/group/diybio?hl=en.

[TRolandB]

Sorry all for delay in reply.

1) We may look at extending the time, but atm biggest issues are extraction / pcr, so wouldn't do much good until after these are fixed.

2) We're not using phenol

3) We have all our reagents from sigma aldrich or nbsbio - you make a good point about these old papers - one thing that always worried us was the melting temperatures of the primers listed in the paper we used - they differ by between 15 and 20 degrees. We initially thought this would be ok due to it being a procedure from a published paper, but perhaps better to go for something that actually makes sense to us.

We are now ordering some primers to detect pea DNA from this paper: http://www.epochlifescience.com/Product/SpinColumn/EconoSpin%20Plant%20Seed%20gDNA.pdf

Hopefully with a more consistent extraction process and new primers we should be able to get results and find out where we were going wrong.  

To unsubscribe from this group, send email to diybio+unsubscribe@googlegroups.com.

[Steve Frese]

If I was going to run those primers, I'd run them at 60C (the paper says 59, which is close enough, but I probably calculated the Tm differently). I dont know how you guys calculate the Tm, but I've had great luck with the quick and dirty "2+4" method (2 degrees for A/T and 4 degrees for C/G: where the sum of the letters = Tm of the primer).  

Extend the annealing and extension steps in that protocol to 30sec each as well (or even 45), to make sure you get the kind of processivity you need to generate your product with the enzyme.

Best of luck

To view this discussion on the web visit https://groups.google.com/d/msg/diybio/-/Sa7XLdYKiRsJ.

To post to this group, send email to diy...@googlegroups.com.

To unsubscribe from this group, send email to diybio+un...@googlegroups.com.

[Nathan McCorkle]

I would start by vortexing a few times rather than the once at the
end. You could also add freeze-thaw cycle(s) if you think lysis is the
problem.

Next I would do an isopropanol (70% v/v) extraction followed by an
ethanol extraction (or 90% isopropanol if you can't get 90%+
ethanol)... then let the tubes air dry for a while until the alcohol
is gone... some people say its hard to redissolve DNA if it dries out
all the way, some say they have no problem... but just dry them long
enough that you don't smell alcohol.

Post-stain gels instead of running them with ethidium bromide, as it
runs opposite DNA so it actually slows down your gel and can decrease
resolution.

Keep us updated!

On Wed, Jun 27, 2012 at 7:12 PM, TRolandB <william...@gmail.com> wrote:
> Hi everyone, our group (London hackspace) have been doing some genetic tests
> (sex typing) recently, with PCR and electrophoresis. We're getting very
> inconsistent results, and think our extraction process is at fault, so we
> want to try some similar techniques but with DNA from foodstuffs, so we can
> be sure of getting the extraction right.
>
> Does anyone have any suggestions of tests you can do that involve PCR and
> electrophoresis of this type?
>
> Will
>

> --
> You received this message because you are subscribed to the Google Groups
> "DIYbio" group.
> To view this discussion on the web visit

> https://groups.google.com/d/msg/diybio/-/3m5IRd20Wt0J.

> To post to this group, send email to diy...@googlegroups.com.
> To unsubscribe from this group, send email to
> diybio+un...@googlegroups.com.
> For more options, visit this group at
> http://groups.google.com/group/diybio?hl=en.

--
Nathan McCorkle
Rochester Institute of Technology
College of Science, Biotechnology/Bioinformatics

[william...@gmail.com]

Great cheers, will report back with results! 

[Nathan McCorkle]

On Mon, Jul 9, 2012 at 1:36 PM, Nathan McCorkle <nmz...@gmail.com> wrote:
> I would start by vortexing a few times rather than the once at the
> end. You could also add freeze-thaw cycle(s) if you think lysis is the
> problem.
>
> Next I would do an isopropanol (70% v/v) extraction followed by an
> ethanol extraction (or 90% isopropanol if you can't get 90%+
> ethanol)... then let the tubes air dry for a while until the alcohol
> is gone... some people say its hard to redissolve DNA if it dries out
> all the way, some say they have no problem... but just dry them long
> enough that you don't smell alcohol.
>

Sorry, to be clear, I meant to do the alcohol extractions after the
chelex steps... since chelex is a chelator, if there are trace amounts
it could be soaking up the Mg2+ in the PCR mix and messing things up
there.

[TRolandB]

We tried the pea primers, but didn't get any result on our gel, so decided to run genomic DNA to test extraction.

Our write up is here.  http://wiki.london.hackspace.org.uk/view/Plant_species_testing

We've done this a few times with different methods, and each time we get a band or smear always running ahead of the 2000 - 100bp ladder, whereas we would expect genomic DNA to run much slower. Methods we've tried are chelex extraction, isopropanol / ethanol extraction, and simple 'homogenisation' (crushing and mixing). Would be great to get some input of possible causes, things we've thought of are:

- DNAses in our reagents

- The fast bands are RNA

>> diybio+unsubscribe@googlegroups.com.