Bioperl: article for Dr. Dobb's Journal

Gustavo Glusman
Sat, 10 Oct 1998 21:46:07 +0200

>I'm writing an article for Dr. Dobb's Journal on Perl and Molecular

Reading it, a few comments -

* Why talk about 21 amino acids, when there are 20? I can understand the
expression 'the 21 letter alphabet', but a phrase like "The other thing to
notice is that certain codons (three in all) don't encode any amino acids."
certainly clashes with "the 21 amino acids"... Why not just talk about 20
amino acids and 3 termination codes?

* "it's typical to work with one subregions" (-> subregion)

* compatability -> compatibility

* "Otherwise, we quickly transcribe the sequence to RNA using Perl's tr()
operator (which actually stands for "translate", not "transcribe") and
return the sequence in RNA form."
-> As a biologist I agree, this is perfectly kosher, but is the kind of
things that inflates the text. For the typical DDJ reader, it would have
been as good to write "otherwise, we quickly translate the sequence to RNA
form using the tr() operator". We're not talking about translating a DNA
molecule into RNA, but about translating its sequence in silico...

* "For historical reasons, DNA subsequences use 1-based indexing,..."
-> For *historical* reasons??

* "there may be small differences between the two sequences, either
because of errors during the sequence determination or because they
are from different but related organisms"
-> Or because they're paralogues, or alleles. (I just suggest relaxing the
'either...or' so it doesn't imply these are the two only reasons.)

* As for the length of the article, I agree that the dp section can go;
maybe just modify slightly the section on align() from "we'll look at this
utility class later" into something like "which is also implemented in
Perl". Conceivably even the section on subseqs could go, but IMHO it would
be more rewarding to trim on the side of what points the article conveys.
I'm just thinking about how many readers will react along the lines of "I
know Perl is good at doing such projects, and the bio stuff confuses me.
Other than claim that large projects can be done using such base classes,
have you really found out that it works well and it doesn't get too
sluggish as the project grows?"

     Gustavo Glusman              Founder/administrator of BioMOO    (public PGP key available)
        Visit BioMOO, the biologists' virtual meeting place, at
     ___________ ___________

=========== Bioperl Project Mailing List Message Footer =======
Project URL:
For info about how to (un)subscribe, where messages are archived, etc: