computational analysis of sequences

R. Jayakumar jakku at
Tue Apr 3 13:10:28 EST 2001

Dear Dr. Tiwari
      From random sequencing, I can understand that you aim to use random or
arbitrary primers to sequence your cDNA clones.  You would also like to find
out what sort of cDNA clones you have and wether they are related to any
known arabidopsis cdNA sequences.  If what I surmised is correct, the rest
of your job is quite easy.  Just go to the BLAST pages at and give your cDNA sequences (after trimming the vector
and the primer sequences; many people do the mistake of forgettting to trim
the vector sequence in which you cloned your cDNA's and that returns highly
ambiguous results) for a BLASTn search against nucleotide sequences.  give
the option as Arabidopsis thaliana in the advanced blasting options to limit
your searching options.   I always found that searching with nucleotide
queries for cDNA's gave quite ambiguous results, since it is the final
translated product that matters.  so I suggest to go for finding the ORF in
your cDNA which will naturally exist and then translate the ORF into the
correct protein sequence using the codon table for eukaryotes and then
submit the protein sequence for a blast search or better still do a blastx
which will take care for all this.  YOu can also try to do a tblastn too
with the protein sequence.  If you do not have tools for identifying frames
and translations, go in for tblastx.  YOu can also try searching for
conserved domains and look for ESTs and STS regions. That should yield
interesting results.
     You may have to do the BLAST search with each of your query sequence
one by one, since as far as I know, it is not possible to do BLAST search
with multiple query sequences in one go.   If you have GCG, then it should
be possible to write some perl script to submit all your query sequences for
BLAST search.  But I am sorry that I don't know PERL or Java. :-)) But
finally the output is going to be for each individual query sequences.
AFter all that is what BLAST is all about.  BLAST is more localised in its
search and so for confirmation of results you can also go in for a FASTA
search which is more global in approach.   Remember, FASTA is more sensitive
for nucleotides than BLAST and gives a better match.
   best of luck..

CSIR- Senior Research Fellow
Dept. of Molecular Microbiology,
School of Biotechnology,
Madurai Kamaraj University,
Madurai - 625 021.
email: jakku at
tel: +91-452-858471-374
office fax(india)= +91-452-859105
Efax(USA): (603)-688-4665
"Life is a comedy for those who think and a tragedy for those who
feel" --Emerson
----- Original Message -----
From: "S.Tiwari" <bsssmt at>
To: <methods at>
Sent: Tuesday, April 03, 2001 10:28 PM
Subject: computational analysis of sequences

> I will be going for random sequencing of my clones from a flower cDNA
> library from a  weed related to arabidopsis. Once I get the clones
> sequenced i have to compare the sequences with the arabidopsis database.
> I would like to know if it is possible to do a blast analysis or
> anything similar with a batch of sequences otherwise if i have to go for
> blast with each of my sequence it will take a lot of time and energy.
> Please help with comment's/suggestions.
> Thanx!
> sumis


More information about the Methods mailing list