strange blast search result - (Mar/13/2006 )

Hello, I am a student working on bacteria (environmental samples). Recently i am using the cyanobacteria primer (359f and 781r). after doing DGGE and reamplification of DG bands, i send purified PCR products for commercial sequencing. But almost all of them give strange result in blast search.
Some samples gave up to 360bp length after sequencing, but they gave like 289/297 (9X%)in blast search, with the beginning 70bp of the sequence un-matched. I guess this may not be the problem in the sequencing coz the peaks (though not sharp) are clear....(?) What could be happening to the 1st 70bp indeed? -- i bet i dont have that many novel genotypes in my samples!!
As i have to do phylogenetic analysis afterwards, so i wish to have longer sequences.

Also, how long should the sequencing output of cyano primers be? some of my cyano give only 300bp long. normally i got ~80% of the actual length for other primers (say, 907r & 341f).

Thank you very much for your help

Fiona

hi

is it neccessary to make a complete match with blast for ur experiment
which website did u blast the sequence or did u just blast it with ur own sequences??

if u tried NCBI and all the rest , it could be possible that ur sequence is not there in the database
or ur sequence got modified during pcr ,,, thats why people usually do sequencing after pcr to make sure there has no error ,, i have heard that sometimes during pcr the original sequence gets modified if the gene sequence is small
i am not sure about it

and it seems not much difference is there in ur blast result ( if its aganist the database) usually only the same sequence from the database will give 100% similarity

hope this clarified some part of ur doubt
but do recheck it since i know just elementary bioinfo

all the best

Hi phytoviridae,

Thank you very much for your suggestion! I've just sequenced some of my samples using both fwd & rev primers & there's sth q.interesting.

I can "knit" the two sequencing result together & i got more or less the same identity fr the blast search, yet, the similarity raised to 37x/38x this time (cool!). Then i pairwise matched the sequencing result of both primers, i found that there indeed sequencing using a single primer is not enough. though each of them gave me 3xx bp reads, it missed out part of the region i expected but gave me some additional but mysterious reads. it's quite misleading that the length of the region it missed out is the same as the additional section, & the peaks are all nice. so originally i thought the sequencing is perfectly ok..

thanks so much for yr advice that pcr with small gene sequence may be problematic & i'm go' to read more abt that ~v~ It is likely that my sequence is being modified in the sequencing reaction (somehow it's a pcr) instead of the previous amplification as the additional section i got fr both primers are not identical.

It's necessary to have longer match as i have to do phylogenies afterwards... otherwise i will be less confident on my own sequences I used NCBI for blast search only.. is there any other places that i can go to?

Thanks again for your help

Fiona

hi

try this site
there is blast for nucleicacid too though it may seem like protein blast



http://www.ebi.ac.uk/blast2/


http://blast.genome.jp/

http://www.ch.embnet.org/software/bBLAST.html

http://dove.embl-heidelberg.de/Blast2/

http://www.expasy.org/tools/blast/

http://www.ch.embnet.org/software/aBLAST.html


http://www.ddbj.nig.ac.jp/search/blast-j.html ( ddbj is japanese , id ot if it wll be useful )

http://flybase.net/blast/( only blasting againist flies )

http://bioweb.pasteur.fr/seqanal/blast/intro-uk.html( general site might be good to go thru it )

actually i tried only NCBI and EMBL blasts till now
try others

good luck
glad could help

Oh thanks for the links!!
i'll try them out~

Fiona ^o^