bacterial microarrays

This forum is dedicated to the use and analysis of microarrays and other aspects of chip technology.

Moderators: mdfenko, Abhijeet Bakre

bacterial microarrays

Postby molecular_micro » Sep 19 2013 1:41 pm

Hi,

I have a very quick question regarding microarray data. I am in no way an expert in this field. I have sent RNA samples to a facility to perform the microarray and undertake the analysis. I was hoping for a simple gene expression profile (ie, relative expression of the genome compared to untreated controls) to be sent back. However, we ran into a problem with signal intensity.

The samples were all OK (RIN was good, yield was good). The array was performed and the facility reported that the signal intensity was very low. Hybridisation controls were apparently good. My query is a simple one. The facility I am using is a generic genomic lab, which typically works with 'higher' organisms, mammalian samples most commonly. Is it likely that the low signal intensity is not in fact 'low' for bacterial microarrays, rather it was an incorrect expectation for signal intensity? My reason for asking is that the contact performing the analysis had noted that low intensity had been seen in some previous bacterial microarrays. Is there high variability between signal intensity and organism?

Owing to my extremely limited knowledge of the physical process of performing a microarray, and the facilities inexperience in undertaking bacterial microarrays, I was hoping someone could shed some light on this matter for me...So I know if I need to get some new chips ordered.

Thanks in advance for your thoughts and advice.
molecular_micro
technophile
technophile
 
Posts: 48
Joined: Jul 16 2012 12:43 pm

Re: bacterial microarrays

Postby mdfenko » Sep 19 2013 6:07 pm

i also have no experience with bacterial microarrays but can you give more information (brand of array, species of array and sample, how weak the signal is, strength of controls, etc).
talent does what it can
genius does what it must
i do what i get paid to do
mdfenko
ModSquad
ModSquad
 
Posts: 1473
Joined: Jan 24 2006 7:39 pm
Location: Staten Island, NY USA

Re: bacterial microarrays

Postby mchlbrmn » Sep 20 2013 7:07 pm

(I also have no array experience, etc... but as an offhand comment....)
Wouldn't bacteria express a smaller number of genes than mammalian, so each transcript should tend to be a more highly represented /ng of RNA, so the signal should be stronger, not weaker? (I assume the array is hybridizing your RNA (or labelled DNA transcribed from it) to DNA on the array.)
mchlbrmn
ModSquad
ModSquad
 
Posts: 3990
Joined: Dec 13 2005 6:03 pm
Location: Boston, USA

Re: bacterial microarrays

Postby molecular_micro » Oct 15 2013 2:04 pm

Hi both,

Thanks for your replies.

I have had the microarray data fully analysed and have found significant changes where I expected....Yay!

I think it was an incorrect expectation of signal intensity by the individual who performed the array and had only cast a brief eye over the data prior to sending data for full analysis.

I have a quick question regarding validation... I am seeing that some individuals/research groups seem to perform qRT-PCR validation while others do not. What is the correct practise here? Also, if one is to perform corroboration assays, what is the correct number of targets to 'check'? There are 50-100 genes across a range of conditions with significantly different expression, it is not viable to qRT-PCR all of these.

If you have any good/core literature to justify the use or non-use of corroboration studies it would be really helpful.

Thanks again for your time and help.

M.M
molecular_micro
technophile
technophile
 
Posts: 48
Joined: Jul 16 2012 12:43 pm

Re: bacterial microarrays

Postby mdfenko » Oct 16 2013 12:02 pm

i don't have specific literature to cite but...

gene expression microarrays are often used to screen for differences from normality.

after finding differences, many investigators will then focus on specific genes with rt-pcr (especially if they are implicated in the condition being studied).

others may just be looking at gestalt differences between tissues and not take it to rt-pcr. they will confirm by repetition (an expensive method, to say the least).
talent does what it can
genius does what it must
i do what i get paid to do
mdfenko
ModSquad
ModSquad
 
Posts: 1473
Joined: Jan 24 2006 7:39 pm
Location: Staten Island, NY USA

Re: bacterial microarrays

Postby molecular_micro » Oct 17 2013 8:58 am

Hi,

Thanks for your response.

In my example I am interested in environmental conditions on an altered phenotype, ie. I found 'x' phenotype (increased resistance to biocide), what is the transcriptional cause of this?. I then microarray'd corresponding RNA during the environmental 'conditioning' to elucidate the cause of increased resistance. Samples were taken n=4. I am extremely close to the end of my PhD studies, with no real time to act upon the microarray data (if time permitted I'd perform the corresponding knock-outs etc.)...So my PI wants me to publish the transcriptomic data alongside the phenotypic data.

I'm not sure in this example, if corroboration is necessary? I understand it would add to the microarray but as you mentioned, if I'm simply interested in the differences accounting for the phenotype (note, I have found alterations where they would be 'expected' based on the phenotypes), do you think corroboration is necessary??

Thanks again for your time.

ps. apologies for my late reply to your initial response, I had not selected for email confirmation of responses.
molecular_micro
technophile
technophile
 
Posts: 48
Joined: Jul 16 2012 12:43 pm

Re: bacterial microarrays

Postby mdfenko » Oct 17 2013 11:24 am

based on what you're saying, you don't have time for proper corroboration. n=4 is certainly not statistically significant. however, if you also ran controls along with your treated samples then you can say that the results you obtained indicate that the changes in expression may be causing the phenotypic alterations.
talent does what it can
genius does what it must
i do what i get paid to do
mdfenko
ModSquad
ModSquad
 
Posts: 1473
Joined: Jan 24 2006 7:39 pm
Location: Staten Island, NY USA


Return to Microarrays

Who is online

Users browsing this forum: No registered users and 1 guest