Arne K Christensen
(by arnec from bio.umass.edu)
Fri Jan 22 14:50:25 EST 2010
I am in the process of designing 4 small riboprobes (20-23nt) that will
be labeled with a fluorophore for use in ISH. Two riboprobes are
antisense and directed against VERY closely related isoforms, and the
other two are their respective sense negative controls. The reasons we
have chosen very small riboprobes are:
1) because the genes are so closely related, there are only a few very
small islands of divergence withing the mRNA sequences that we may target.
2) because we are not really set up for molecular work, it may be
cheaper to order the labeled synthetic RNA sequences rather than
transcribe our own.
The problem I am running into is that for any 20-25nt sequence I select,
there is always a contiguous ~50%-70% of the sequence that is identical
to other genes. The selected sequences are so short, the chance of a
redundant sequence anywhere in the genome of the species I am working
with is very high. To make things worse, I am working with a non-model
species, and more than half of the transcriptome is not getting screened
with a BLAST.
I would greatly appreciate any information people have regarding the
selection of small ribroprobes for ISH. Is 50% ID with other sequences
divergent enough to get specific signal?
More information about the Methods