Jump to content

  • Log in with Facebook Log in with Twitter Log in with Windows Live Log In with Google      Sign In   
  • Create Account

Submit your paper to J Biol Methods today!
Photo
- - - - -

I'm working on an automated DD2CT value generator for Life tech qPCR machine


  • Please log in to reply
2 replies to this topic

#1 keith_neuro

keith_neuro

    member

  • Members
  • Pip
  • 2 posts
0
Neutral

Posted 24 April 2014 - 09:18 AM

Hi all,

I'm trying to cut down the time and effort that we spend  cleaning data for qPCR. 

 

A brief background, I was a neuroscience post-doc who spend a lot of time running gene expressions studies in an epigenetics lab.

Since then, I realized that a lot of the stuff that we did with qPCR can be easily automated, made more user friendly, and made easier to share with colleagues. 

Also I don't believe that we should pay thousands of dollars for software that, to be honest, wasn't very good.

 

So I created a website called www.stirplate.io which is a free platform for scientists to store, share and analyze their data. 

 

I started working on qPCR first as it was near/dear to my heart and I'd love to do some testing with people if they are interested (there is no cost involved, so I'm not trying to sell anything here). 

 

So, here is what I have working:

Drag/drop your raw output and a "treatment code" file (so I can decode your sample names) and I can get cleaned results in about 15-30 seconds per plate.

Here is what is done:
1. All samples are merged with the treatment codes and organized by treatment. 

2. I check the CT-SDs to make sure the duplicates/triplicates are good. For any triplicate group with an SD of 0.2, I remove the outlier sample (and report that to you).

3. DD2CT values are generated by subtracting the endogenous control (you choose the control), from the GOI (DCT), then an average of the DCT from a control group (you choose) is taken to generate the DDCT value. The DD2CT value is generated from that number.

 

4. I provide 2 graphs and 1 table. A bar graph w/ the means by group, a scatter plot so you can see those data plotted, a summary table (N per group) and which samples were omitted due to not reaching threshold or due to being a bad triplicate.

 

5. You can interact with the graphs: For example, if you see an outlier in the Scatter plot, you can click on that sample, make a note as to why you are removing it, and the graphs re-write themselves in about 2 seconds. The same applies to changing the endogenous control, or control group, it only takes about 2-3 seconds.

 

6. You can export just the cleaned values via excel from there.  

I made a 3 minute  video here or you can sign up on the site and give it a shot. 

 

I have this working quite well for the Life Tech Step 1, and the Viia 7 (using the microfludic cards).

 

I'm also working on the Life Tech 7500 and ABI 7900 right now.

 

 

Again, this doesn't cost anything, I'd just like to get feedback from the qPCR community to see how I can improve on this technique. I'm building tools like this for  a lot of different techniques soon, but I'd like to nail down qPCR initially.

Thanks all!
Keith

 



#2 Trof

Trof

    Brain on a stick

  • Global Moderators
  • PipPipPipPipPipPipPipPipPipPip
  • 1,190 posts
106
Excellent

Posted 25 April 2014 - 12:43 AM

Just curious.. ABI doesn't have build-in software app, that will calculate you relative-quantification, just by selecting calibrators and references and so? Or has a separate one, that is very expensive?

 

Because, RelQuant module is (if I'm not mistaken) included with basic LightCycler 480 software, works directly inside the software suite, settings are within other plate settings and so on.

It's pretty much helpful for realtive quantification, in case of single calibrator within run, it all samples pairs automatically, supports multiple reference genes (even with references on different plate), supports calculations using real reaction efficiency, and has several options you can choose from. Doing quantifications without it would be a pain in the ass.

 

I would found it really bizarre, that the biggest competitor doesn't offer at least a part of this in the basic offer.


Our country has a serious deficiency in lighthouses. I assume the main reason is that we have no sea.

I never trust anything that can't be doubted.

'Normal' is a dryer setting. - Elizabeth Moon


#3 keith_neuro

keith_neuro

    member

  • Members
  • Pip
  • 2 posts
0
Neutral

Posted 26 April 2014 - 09:26 AM

Hi,

They actually do provide that, but the set up process is tedious and not very flexible. 
For a 384 well plate, it used to take just as long to set the plate up as it did to run it. It's much like other science software that we use, the user experience is just poor and people  (myself included) tended not to use those features. To be honest, ABI/Life tech's software is just outdated and not really easy to use. (hence why they still ship Windows XP machines).

 

 

Furthermore, sharing that information with other people is difficult. I'd have to send the raw file (.eds file) which means you have to have their software to open the file. 
It only works on PCs, doesn't work on 64 bit OS etc etc. 

That whole process is much easier when things are built on the cloud.
 

On Stirplate, you would just export your results, make a treatment file (sample name & treatment) and just drag/drop both files into Stirplate. That's it. Any changes, are as simple as a click, and anyone you share it with can see the changes that you made, when you made them, and why. 

 

Edit: If you would like to try it, I'll send you some of my raw output and you can give it a shot. Happy to send you a link to the ABI software as well and you can try it side/side to see what I'm talking about. 


Edited by keith_neuro, 26 April 2014 - 09:26 AM.





Home - About - Terms of Service - Privacy - Contact Us

©1999-2013 Protocol Online, All rights reserved.