Skip to content

new book

5 messages · Robert Gentleman, Martyn Plummer, A.J. Rossini

Original
Robert Gentleman
# 
Just thought I would mention that there is a new O'Reilly book out,
"Developing Bioinformatics Computer Skills", which is ok, pretty
superficial about some things but at least it lets you know what is
happening and where.
R gets a few pages in there, 394-396, mostly nice press and accurate
except that Bill Venables (hi Bill) may be surprised to find out that
he is a member of the core development team.
  Robert
ps: they also have an Xgobi reference
Martyn Plummer
#
On 30-May-2001 Robert Gentleman wrote:
This came up a couple of weeks ago on Slashdot (which I should really
stop reading).  I got the impression from the ensuing discussion that
it was targetted at biologists,but if you found it useful I shall pick
up a copy.  In the same discussion, the monograph "Biological sequence
analysis: probabilistic methods of proteins and nucleic acids" by Durbin
(ed) was highly recommended by some people.  Do you have this?

Here is a link to the O'Reilly site which contains contents, and a
sample chapter.

http://www.oreilly.com/catalog/bioskills/

I remember buying an O'Reilly book called "The whole internet: a users
guide and catalogue". I wonder if this one will have an even shorter
shelf life.

Martyn
-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-
r-devel mailing list -- Read http://www.ci.tuwien.ac.at/~hornik/R/R-FAQ.html
Send "info", "help", or "[un]subscribe"
(in the "body", not the subject !)  To: r-devel-request@stat.math.ethz.ch
_._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._
A.J. Rossini
# 
MP> This came up a couple of weeks ago on Slashdot (which I should
    MP> really stop reading).  I got the impression from the ensuing
    MP> discussion that it was targetted at biologists,but if you
    MP> found it useful I shall pick up a copy.  In the same
    MP> discussion, the monograph "Biological sequence analysis:
    MP> probabilistic methods of proteins and nucleic acids" by Durbin
    MP> (ed) was highly recommended by some people.  Do you have this?

    MP> Here is a link to the O'Reilly site which contains contents,
    MP> and a sample chapter.

    MP> http://www.oreilly.com/catalog/bioskills/

    MP> I remember buying an O'Reilly book called "The whole internet:
    MP> a users guide and catalogue". I wonder if this one will have
    MP> an even shorter shelf life.

I commented privately, but since this comes up, I'll comment again: if
you know your way around genetic programs, perl and bioperl, and
understand general scripting tools, and have a smattering of
informatics knowledge, the book isn't too useful.  I've got a
protein-folding friend who plans on using it to teach a comp bio
course in microbiology; he considers the skills mentioned in the book
to be the bare-minimum for that, though it's short on theory.

The Durbin book is a nice intro to current state of the art DNA and
protein sequencing methods, but not much more (of course, it lives up
to its title).  That being said, I found it incredibly useful for
seeing what has been done recently.  YMMV.

I think part of the problem, and this is seen by those who recommend
the Durbin book, is that there are a number of different ways of
looking at the "bioinformatics" problem.  Problematically, you've got
something like:

A. Large quantities of data, and handling, storage, backup
B. Annotations and external (data) linkages
C. Computational Biology
D. Statistical issues (experimental design/analysis and exploratory
   approaches; 
E. Combinations of the above two, including such topics (one of many
   examples): stochastic DE modelling using mixed effects models.

and scientifically, you've got one or more of:

1. sequencing (DNA and/or protein)
2. protein structure (secondary/tertiary, based on the above)
3. gene expression arrays of many forms (cDNA, membrane, gold,
   oligios). 
4. SNPs 
5. genetics and evolutionary biology

and then there is the application area (cancer, HIV, other chronic
diseases; comparative studies and animal/plant/non-human organism
models, etc).

The classifications that I use are even up for grabs, depending on who
you are and who you talk with; it's just 2 possible views of this high
buzzword content area, sigh.  I know I've left out a few areas, here.

best,
-tony
Robert Gentleman
#
On Wed, May 30, 2001 at 07:31:48AM -0700, A.J. Rossini wrote:
I agree, but for those who are just getting started -- especially
   statistics grad students (or biology grad students or comp. sci) it
   can be very useful.
   I am actually a bit tempted to use it as a text for 1st year
   MSc/PhD's here, intro to stat computing. At least it gives them
   some examples that are relevant (and most know no Unix and no Perl
   and no....)
Yes, and Gusfield -- Algorithms on Strings Trees and Sequences is
  reasonable for what it does as well.
I know very little about B. and found that DBCS at least gives some
  idea of what is available, what it looks like and how you might
  begin to deal with it.

  Yes, the field is developing so rapidly that it is useful now and
  won't be as good in 6 months and probably will be useless in 12 or
  18, but then so will the machine I'm typing at.

  
    

      
      
    

  


  


      

    

    

      
      

        


  

        

      A.J. Rossini
    

    

      
      #
    

  


  

          

      
      
      
    

        
>> I commented privately, but since this comes up, I'll comment
    >> again: if you know your way around genetic programs, perl and
    >> bioperl, and understand general scripting tools, and have a
    >> smattering of informatics knowledge, the book isn't too useful.

    RG>    I agree, but for those who are just getting started --
    RG> especially statistics grad students (or biology grad students
    RG> or comp. sci) it can be very useful.  I am actually a bit
    RG> tempted to use it as a text for 1st year MSc/PhD's here, intro
    RG> to stat computing. At least it gives them some examples that
    RG> are relevant (and most know no Unix and no Perl and no....)

To clarify, I would agree that this is a good use of the book.  That
isn't the audience that I described it as "not very interesting for". 

However, a good number of people on R-devel are beyond that level, in
most respects (if this had been R-help, on the other hand...).

best,
-tony