Workshop feedback and future plans

Thanks for the feedback and gracious comments on the course, website and instructors. Here are some of the ways we are thinking of implementing your suggestions. Please let us know if there is anything we missed or can improve further on.

For everybody who asked about how long the website will be up - we have no intention of shutting it down. However, the website may evolve as we incorporate your suggestions and lessons learnt from conducting similar workshops. If for any reason the URL has to change, we will send emails to all previous workshop participants. If you are really worried, you can download a snapshot of the website as of 1 June 2012 at the bottom of the references page under Archival material.

Please do continue to email cliburn.chan@duke.edu if you have any more suggestions on how to improve the workshop, or if you have any further questions related to the workshop.

Pre-course exercises

This is a great suggestion for how to get workshop participants on a more equal footing before starting the course. We really struggled with trying to keep the more experienced participants engaged without losing complete newcomers to computing. We will incorporate the following pre-course exercises in the future.

1. Basic familiarity with the Unix terminal - opening a terminal and the basic commands for navigation e.g. ls, pwd, cd, cp and mv.
2. Open a text editor, load a file downloaded from Data Samples, turn on soft wrap and do a simple Find and Replace operation (i.e. no regular expressions).
3. The difference between an ipython session and running a python program. Start an ipython program, use the built-in help and print Hello world. Open a text editor, save the program consisting of print "hello world" as hello.py and running it on the terminal with python hello.py.

More real data samples

We hesitated with giving too much real data because real data tends to be messy, and we wanted to avoid overloading course participants with specific details before you’ve nailed down the basic concepts. Having said that, it was clear that real data was a great motivator, and provided a context where you could see how the concepts learnt would be applicable to your situation. So we’ve decided to create 3 levels of “real” data for future courses - working through these increasingly challenging sets should be a great learning experience.

1. Totally fake data but in a realistic context similar to the microarray expression data
2. Real data, but pre-cleaned to remove irregularities
3. Real, real data with all its faults and inconsistencies

Review sessions

Another great idea to consolidate new concepts. We will begin each day with a 15-30 minute review of the previous day’s work for consolidation, and end the day with simple (hopefully) exercises to do at home.

Answers to exercises on website

We will prepare answers to all the exercises and create a new Solutions to exercises page.

New workshops

There was unanimous support for the question “Would you be interested in attending more similar workshops in the future?”, which is extremely gratifying. We must have done at least some things right :-). The most popular suggestions were:

1. More Python for biologists with emphasis on graphics
2. More BioPython
3. Statistics for biologists
4. Data management and database course
5. More Inkscape

One possibility is that we run two workshops a year - a general introductory workshop in summer (PCfB) and a more specialized workshop in winter that may vary from year to year and covers the topics suggested. We will keep you posted to see if there are sufficient resources to make this possible.

Provide coffee

We will try our best to apply for NIH, NSF and Gates Foundation funding to cover this expensive but essential resource!