Python and C

In early October 2006 I lectured at the University of Stellenbosch on how to make C and Python work together, various ways to parse files, and more.

• ctypes, calling C functions and passing a callback; example is based on a numerical integrator
• C extension, calling C functions through Python's C extension interface. Uses evaluating a point for Mandelbrot set membership as the driving example.
• SWIG example
• Using ElementTree to extract data from an SBML file
• Parsing a file "by hand" with regular expressions
• Parsing with PLY
• The glory that is threads

Programming for Bioinformatics in Python

At the end of July 2006 I taught an intense two week NBN course titled "Modern Web Applications Development for Bioinformatics" and based on the TurboGears framework. Here are the lecture notes.

• Getting Started with TurboGears
• The IBM developerWorks article on TurboGears
• Loading data
• Version control
• Creating a taxonomy server
• Editing the taxonomy database
• Logins and authentication
• Uploading files
• Job management
• ElementTree
• For CSS I'll use other resources
  
  • Online demos
  • Wikipedia on CSS
  • W3C page on CSS
  • the Zen CSS Garden
• Javascript validation
• Ajax
• Ajax example

Programming for Bioinformatics in Python

In August-September of 2005 I taught an intermediate level programming course at the NBN (National Bioinformatics Network) in South Africa. Here are the various notes and assignments for the course.

• Parsing FASTA files, Assignment #1 (data files simple.fasta, fasta_reader.py, br_sequences.fasta, swissprot.dat)
• BLAST output parsing, Assignment #2 (data files blastp.txt, blast_parser.py)
• BLAST searching
• Making hydrophobicity plots (data files bacteriorhodopsin.fasta, fasta_reader.py, data/savitzky_golay.py)
• The Monday (22 August) lecture is on chemical nomenclature and will cover my essays
  
  • Drawing molecules
  • Exceptions to the graph model
  • Naming molecules
  • Naming known molecules
  • Naming novel molecules
  • Molecular Formula - History of Chemical Nomenclature
  • Systematic Name - History of Chemical Nomenclature
  • Paper databases -- History of Chemical Nomenclature
  • WLN -- History of Chemical Nomenclature
  • Computers -- History of Chemical Nomenclature
  • Chemical Informatics Toolkits
  Afterwords I'll describe the SMILES molecular line notation and its extension to molecular reations using SMIRKS. I'll describe the subgraph matching problem and describe SMARTS patterns.
  The exercises will be based on the on-line demos available from Daylight and OpenEye. The network is down now so I'll have to work on it during the breaks tomorrow.
  
• Assignment #3 (chemical informatics)
• The Tuesday (23 August) chemical informatic lecture will cover ways to compare two molecules, with an emphasis on molecular fingerprints. This will lead into a discussion on chemical databases.
• The Wednesday Python lecture – Parsing GenBank
  Assignment #4 (Python) (data files br_plot.py, bacteriorhodopsin.fasta, br.gp, 1Z9KA.gp)
• Assignment 5 (chemical informatics)
• The Wednesday chemical informatics lecture will cover the PyDaylight and OpenEye toolkits
• Some information about chemical databases which examples taken from gNova's CHORD extension to PostgreSQL.
• A short lecture on list comprehensions
• Usability, day 1, August 29
• For the Python lecture for August 29 I'll talk about Python generators. I'll make sure everyone has CherryPy installed on their machines and walk through the tutorial. More advanced programmers may want to read the IBM developerWorks article as well.
• A good tutorial on HTML. I like it better than the HTML tutorial I started doing in class. I hope you get to the section on HTML forms!
• User requirements for the similarity project
• Using Python to write an HTML table for GenBank sequence features
• Assignment 6
• Iterators and more FASTA parsing
• HTML forms
• More on web programming with CherryPy
• Assignment 7
• XML
• XML-RPC and web services
• Data clustering
• Student teaching schedule

---

Introduction to Programming for Bioinformatics in Python

In February 2004 I taught an introductary programming course at the NBN (National Bioinformatics Network) in South Africa. It was part of an intense and impressive 7 week training session for bioinformatics research with topics including bioinfomatics theory, algorithms, databases, software, unix, programming and even grant writing.

The portion I taught was two hours a day, six days a week for the first two weeks. I started off with a translation of a introductory Perl course Lincoln Stein taught but after the first day I realized that most of the people in the class had very little programming experience so I started the next day with the basics and lots of slides.

Some of the students already knew how to program so I made a wide range of exercises with the expectation that the beginning students would only get through the first few in each problem set while the most advanced ones would still be challenged with the harder ones. Judging from the feedback it worked out pretty well.

Here are all the slides in PDF and PowerPoint format. I made them using Apple's Keynote but that format is verbose and requires directories so too much work to put it on-line. Contact me if you need them.

• Introduction to Python [html]
• Strings in Python [PDF|PPT]
• Variables and References in Python [PDF|PPT]
• Lists and the 'for' loop [PDF|PPT]
• Stepping through a 'for' loop [PDF|PPT]
• The if statement and files [PDF|PPT]
• Working on exercises (a few notes first) [PDF|PPT]
• Code Blocks and Indentation [PDF|PPT]
• Functions [PDF|PPT]
• Why? [PDF|PPT]
• Dictionaries [PDF|PPT]
• Advanced Exercises [PDF|PPT]
• Sorting and Modules [PDF|PPT]
• Searching and Regular Expressions [PDF|PPT]
• Important modules: Biopython, SQL & COM [PDF|PPT]
• "Everything Else" [PDF|PPT]
• Answers to all the exercises

Some of the exercises used data files. I think this is the complete list. If missing I note why.

• 10_sequences.seq
• ambiguous_sequences.seq
• ls_orchid.fasta
• many_sequences.seq (it is all of SWISS-PROT in FASTA format)
• sequences.seq
• weights.txt

I really enjoyed teaching the course. It was a lot of work but the result was seeing many people really get into it and faces light up with understanding and achievement.

To the whole class, best wishes and good luck for the future!