Skip to Content

Coding Problem 6 Now Defined in Scripting Languages/Bioinformatics WIKI

In the "Scripting Languages and Bioinformatics" WIKI, I have completed a new page here:

https://wiki.sdn.sap.com/wiki/display/EmTech/Bio-InformaticBasicsPartII-AlignmentToolsforRelatingProteinGenestoProteinPrimaryStructures

and also defined Coding Problem 6 at the end of this page.

This problem involves scripting an interactive session with a web site and parsing what the web site returns at various points duing the session.

Am looking foward to a script that solves this problem, if anyone feels like spending the time to do it.

Add comment
10|10000 characters needed characters exceeded

  • Get RSS Feed

3 Answers

  • avatar image
    Former Member
    Aug 06, 2008 at 06:04 PM

    Hello.

    Following will be the (fully functional) sketch of the solution scripted in PHP. While I start to write this I notice that there is a glitch in the script - it returns a DNA sequence, but not the correct one as it seems.

    Anyway, I continue posting it as it is because the whole code is a nice example on how to utilize a given functionality designed to another user interface and tailor it to your needs. the same way like we interact with the given 'program' we could use a stock exchange website to automate some of our tradings or we could write a 'bot' to collect SDN points by analyzing questions, submitting them to other search engines and returning the result to the original poster.

    To better understand how this works I encourage the reader to download a network sniffer, e.g. [Wireshark|www.wireshark.org], use it to analyze the network traffic during the regular procedure in the browser, additionally analyze the source code in the various screens of the given website and finally find out what the site requires and what it returns. Comparing this with the rather trivial script parts where the data to be submitted is put together should easily allow you to help me find the error and yourself learn how to script against an arbitrary website. Apart from that I spend a beer at the next possible occasion to the one who finds the bug in the original script 😉

    Okay, before we go into the details, I want to mention that this script can easily be reprogrammed in ABAP (at a more recent release where regexes are available) using the regex an http_client classes. The same of course works in Java. Maybe someone else feels encouraged to give it a try.

    Have fun,

    anton

    to be continued ...

    Add comment
    10|10000 characters needed characters exceeded

    • heh heh heh.

      Point taken - I'll add the tag myself !!!!

      And I will try to find the multi-section return from tblastn as soon as possible.

      Otherwise, I'm afraid you'll tell me I have to write that loop also! (Just kidding, just kidding!)

      Best

      djh

  • Aug 06, 2008 at 09:57 PM

    Hi Anton and David,

    I really like this collaboration around a bioinformatics problem. May be one of these days I find a quiet minute and try out the solution.

    Thanks to both of you for doing that, Mark.

    Add comment
    10|10000 characters needed characters exceeded

    • Mark -

      Thanks much for the kind words.

      As Anton has pointed out, what we'll eventually be doing is developing a framework for WDA->PHP calls that could be useful to traditional SAP customers in a variety of ways.

      Best

      djh

  • May 05, 2009 at 02:21 PM

    Anton - did you delete your php code from this thread ???

    I can't find it anywhere.

    Add comment
    10|10000 characters needed characters exceeded