Favorite Tools for Aligning Multiple Sequences

36 views
Skip to first unread message

L

unread,
Jun 25, 2012, 8:27:31 PM6/25/12
to diy...@googlegroups.com
Hey there everyone, 
I'm looking for a way to quickly create protocols for transformation of artificial yeast chromosomes. Got lots of ideas, the means to try them all, and now I'm just looking into what restriction enzymes I'll need. Confused by BLAST, any help using that would be awesome. 

Thank you! 

Nathan McCorkle

unread,
Jun 26, 2012, 12:46:10 PM6/26/12
to diy...@googlegroups.com
When I was doing this last year, I downloaded the list of REs from
promega and NEB, then did a set-intersection with those lists and the
vector MCS+RE sites NOT PRESENT. That gave me a list of enzymes that I
could use for engineering, and I then just sorted by price to pick the
cheapest ones.
> --
> You received this message because you are subscribed to the Google Groups
> "DIYbio" group.
> To post to this group, send email to diy...@googlegroups.com.
> To unsubscribe from this group, send email to
> diybio+un...@googlegroups.com.
> To view this discussion on the web visit
> https://groups.google.com/d/msg/diybio/-/lEbeFHPh7OoJ.
> For more options, visit https://groups.google.com/groups/opt_out.



--
Nathan McCorkle
Rochester Institute of Technology
College of Science, Biotechnology/Bioinformatics

Cory Tobin

unread,
Jun 26, 2012, 8:01:07 PM6/26/12
to diy...@googlegroups.com
If you're trying to find restriction sites in a sequence, this works
well: http://www.restrictionmapper.org/ And if you are trying to
automate it, they offer the code here:
http://www.restrictionmapper.org/automate.html It's in Perl.

-cory

L

unread,
Jun 27, 2012, 2:04:03 AM6/27/12
to diy...@googlegroups.com
Awesome, this is very simple to use! Do you have anything for aligning multiple similar sequences of several similar organisms by any chance? 

-L 
 

Cory Tobin

unread,
Jun 27, 2012, 3:24:08 AM6/27/12
to diy...@googlegroups.com
>> Awesome, this is very simple to use! Do you have anything for aligning
>> multiple similar sequences of several similar organisms by any chance?

ClustalW
EMBL has a web interface for the algorithm here:
http://www.ebi.ac.uk/Tools/msa/clustalw2/

Also, BioPerl and BioPython both have a clustalw module.
http://doc.bioperl.org/releases/bioperl-1.0.1/Bio/Tools/Run/Alignment/Clustalw.html
http://biopython.org/DIST/docs/api/Bio.AlignIO-module.html
It's a fairly common algorithm, so any decent bioinformatics package
is likely to have it.


-cory

shamrock

unread,
Jun 27, 2012, 8:48:55 AM6/27/12
to diy...@googlegroups.com
Hi L,
 
I didn't quite understand your question-are you looking for protocols to transform yeast cells with the artificial chromosome or are you looking for protocols for cloning large pieces of DNA into artificial chromosome vectors?

Tristan Eversole

unread,
Jun 30, 2012, 5:22:23 PM6/30/12
to diy...@googlegroups.com
Alright, so you're doing multiple sequence alignments? Like, global alignments?

Finally, a chance for me to be useful. I don't do synthetic biology, but I am interested in evolution, and, unsurprisingly, multiple alignment is very, very important to molecular phylogenetics-- if your alignment is no good, your phylogenetic reconstruction will be no good. However, picking an alignment method isn't that easy; "The Phylogenetic Handbook" lists, IIRC, some 43 different alignment algorithms. They take fundamentally different approaches, too; my favorite, which seems somehow thematically appropriate, is that of genetic algorithms.

"The Phylogenetics Handbook" has this to say about Clustal:

> Although ClustalW is still the most popular alignment tool, several more recent methods have now been shown to perform better in terms of accuracy and/or speed.


It has a nice table under the "Which program to choose?" section. You're probably in the realm of the first case:

> Input data: 2-100 sequences of typical protein length (maximum 10,000 residues) that are approximately globally alignable
>
> Recommendations: Use ProbCons, T-Coffee, and MAFT or MUSCLE, compare the results using AltAVisT. Regions of alignment are more likely to be correct. For sequences with low percent identity, ProbCons is generally the most accurate, but incorporating structure information (where available) via 3DCoffee (a variant of T-Coffee) can be extremely helpful.

Personally, I tend to use T-Coffee. The tradeoff here is between accuracy and computational requirements: T-Coffee is generally more accurate, but Clustal has lower memory requirements. I think Clustal is currently in the realm of "everyone uses it because everyone else uses it".

I'm no expert, either, so take this with the requisite grain o' salt.

--T.
Reply all
Reply to author
Forward
0 new messages