Massively clone gene homologs

[Phuong]

Hi all,

I am working on a multi-member family of cysteine-rich peptides in plants. We already obtained a few genes but in a very inefficient manner. First, I design the primers based on peptide sequences and clone the full transcripts. Then, from the newly found transcripts, I design the primers against the conserved regions such as signal peptides and 5' untranslated region to randomly clone the homologous genes. However, I normally can get only the abundant genes, leaving a lot of low-abundant or non-expressed genes undiscovered.

Now my professors wants me to obtain a large number of these genes. I don't know which method I should follow to get that. I really hope I can get some hints from you. I appreciate any comments on my work.

Thank you.
Phuong

[relaxin]

For low abundant cDNA or non-expressed genes, you may have to screen genomic library with labeled cDNA that you already have as probes. Separate the clones based on the intensity of the hybridization signal with probes of different regions of the cDNA (5'-probe, mid-region, and 3'-probe). Identical clones will give strong signal, and distant homologs will give weaker signal. Characterize the clones by restriction digestion and DNA sequencing.

[Phuong]

Hi Relaxin,

Thank you so much for your reply. I am searching the literature to find the detailed protocol. Hopefully I can succeed.

Merry Christmas to you and your family ^_^

Phuong

[Onyourcase]

As a first step I would do a Southern Blot to check how many genes there are in any given genome. You can also use this to help you to size select fragments to clone.

[mchlbrmn]

If you have access and/or money, how about the new deep sequencing techniques, like Ilumina or 454? That should let you see the less abundant along with the abundant genes (or even sequence most of the genome if you had the time/money, I guess).

Another idea is to prescreen the clones in the method you are currently using if you have a surplus of clones and want to remove the repeat clones. You could dot the clones in an array on membrane and stringently hybridize with the most abundant clones, or you could do qPCR with specific primers for the most abundant clones.

I've got to go. I'm not sure how well thought out this was, but there are some ideas.

[relaxin]

What is the plant you are working on? Chances are its entire genome has been sequenced. You can simply search the data base with your cDNA sequence and find the homolog sequences. No need to screen genomic library.

[Phuong]

Hi,

Thank all for your suggestion and advice. I will think of them carefully when designing my next experiment.

@Onyourcase: I tried Southern blot before with a probe prepared by PCR against the most abundant template. The probe is about 600bp. However, I only see a few bands of the ladders and none of my DNA. I don't really know what went wrong back then.

@mchlbrmn: I am currently in Singapore, and as far as I know, the new deep sequencing services are not handy here. So I still have to stick to the old method of sequencing.

@Relaxin: I am working on a few plants belonging to Rubiaceae and Apocynaceae family. I already checked last time and found that their genomes have not been sequenced.