HUB Training - initial proposals

From HUB
Jump to navigation Jump to search

Training people in Heidelberg in bioinformatics is another way we (HUB) could contribute to bioinformatics in the area, have a chance to work together on a project, and make people more aware of our activities.

An initial suggestion would be e.g. three days introducing programing in the context of some NGS-specific data analysis problems.

If you'd like to meet and discuss setting something like this up, add your name to the list below.

Initial meeting agenda (26th June 2013)

What to teach?

Initial ideas for topics:

  • NGS software
  • Linux command line
  • Programming (e.g. in Python)
  • Maybe tie these ideas in together to do an intro to command line and programming, and take the examples from NGS analysis?

How to choose a topic?

  • People attending first meeting
    • quickest, but might miss what people really want
  • Survey of HUB
    • not so quick, but allows us to get more input from the community, which is a key HUB thing, and we can also ask already for volunteers to teach particular topics
      • Aidan's choice
  • Survey non-HUB people in our institutes/departments
    • takes longest to organise, I guess, and hard to do in a systematic manner


For now, our ideas on:

  • Possible venue(s)
    • Maybe try several different venues for course? Get's people seeing more of the relevant institutes etc. [AB]
  • Length, timing (e.g. just mornings? weekends?)
  • Trainees bring own laptops, or local machines?

Minutes, 26th June 2013, Star Coffee Bismarkplatz

Present:Alice Krebs, Aidan Budd, Matt Betts, Grainne Kerr, Jon Fuller, Holger Dinkel

We discussed:

Types of possible audience:

  • undergrads
  • master's courses
  • phd students
  • "researchers":
    • bioinformaticians
    • bench scientists who want/need to use computers more in their work
      • strong preference for this (bench scientists) as logistically (finding times, negotiating with administrations of differnet programs) assumed to be the simplest and it's what we're used to teaching
      • can of course consider alternative audiences in the future


  • Strong preference for just one day
  • Can of course consider long ones in future, try one day as an experiment

Weekend or weekday:

  • At the weekend
    • Advantage is that people don't have to forgo work time/expeiments/classes to join
    • Disadvantage that access (public transport, access to buildings) to venue, providing food, can be a problem at some venues
  • During the week
    • Advantage is that people don't have to forgo fun/family time to join

Our clear preference, as potential organisers/trainers, was for during the week, we'll look at seeing if there might possibly be future interest in something over the weekend

Location: To be decided at future meeting

Which computers to use:

  • people bring own laptops
    • advantage:easier for people to work with system when they leave the class
    • disadvantage:lots of time can be wasted trying to get things working on these machines
  • provide pre-installed machines locally
    • our preference (as we feel that the disadvantage above outweighs the advantage)
    • consider setting up a HUB clinic somewhere for people to come (at some point, somewhere) to get help from experienced HUBers getting software working on their machines - maybe poll interest of this at HUB7?

When to try and run it:

  • autumn 2013

Things to bear in mind while planning it:

  • make it HUBy; small discussions, try some pair programming, insist on going out afterwards
  • maybe get some t-shirts ready for then
  • use the HUB wiki for the training pages, so that people see the resource and are introduce to HUB
  • Alice happy to be a guinea pig and then a TA


  • 1 day course (if goes well, can consider repeating and/or doing a follow-up/more advanced course)
  • during the week, not at weekends (maybe give people chance to register interest, but to say they could only come at weekends, so that we judge whether or not in the future we should try/consider that)
  • topic:Introduction to the Linux commandline for NGS analysis
  • make it HUBy if possible
  • do it in Autumn 2013
  • hold a follow up meeting to:
    • select possible dates
    • put together provisional program
    • divide up responsibility for different aspects of running the course
    • choose possible venues

Second planning meeting agenda

Thursday 4th July 2013, 08.00, Star Coffee, Bismarkplatz HD

Present:Aidan Budd, Alice Krebs, Matt Betts, Maria Secrier, Jon Fuller, Adam Gristwood, Grainne Kerr

Agreed so far

We reviewed the decisions made at the previous meeting concerning the course, and the reasons for making these decisions:

  • 1 day course (not longer)
    • less effort for us to organise for this first attempt at organising a course together
    • easier for people to take one day off from their research than 2+
  • Use pre-installed machines, not doing exercises on laptops brought by trainees
    • this avoids wasting time attempting to solve OS-specific problems during the course
    • it also makes it easier for us to ensure that all trainees will be working with a machine which has the necessary software installed on it
  • hold the course on a week/working day (Monday-Friday)
    • the organisers prefer not to spend their weekend time on this
    • much easier to get help, support, logistics, access during the week
  • topic: Introduction to the Linux commandline for NGS analysis
  • make it as HUBy/unseminary as possible
  • Autumn 2013

Target Audience

We spent lots of time discussing and trying to decide on our ideal audience for the course. We attempted to decide between a course aimed at either:

  1. people with no Linux command-line experience at all
  2. people with some experience using the command line

If we go for people with some experience, we want to include as part of the application form, a set of questions aiming to very specifically test their competence, so that we can target the material as specifically as possible at the applicants. For example, we'll ask "Write down in one line how you would move a file titled "test" from your current directory to your home directory" (in this case 'mv test ~')

However, we decided we didn't want to decide on the audience already today. Rather, we think it best for us to decide on one or more specific examples of 'simple' analyses one would do using the command line in NGS analysis, then look at the commands needed to do this, and then teach aiming to deliver by the end of the day of teaching, them to the level of competence of doing that themselves, and understand the concepts involved well enough to be able to address a similar problem (i.e. needing a similar level of understanding of how Linux works) but using different tools/examples. We will then consider what level of competence the "average" person would need to have, to be able to learn this by the end of the day, and put that level as the prerequiste for attending the course.

Matt and Grainne agreed to put together some examples of this from their work, so that we can meet again in July to agree on the level needed as a prerequisite for the course.


We agreed to try for the BioQuant teaching room, as it is most accessible of the places we work (i.e. not on the top of a hill as is the case for EMBL and HITS). Matt will look into checking whether this would be OK, and if so, when in Autumn i.e. in September/October/November 2013.


  1. Prepare examples of full set of Linux "commands" used to do some "simple" NGS analysis using the command line: Matt and Grainne
  2. Find a date for third planning meeting: Matt?
  3. Check possible availability (and whether we would be OK to use) BioQuant training room in Autumn (September/October/November) 2013: Matt

Third planning meeting

Tuesday 30/07/2013, 7pm, Hemingways.


  • Go through example analyses, decide on common points and what we can usefully teach in a day.
    • Please prepare examples of this in advance. Better still, add them to the discussion page before we meet.
  • Draft the following:
    • student objectives - what they can expect to learn / be able to do by the end of the day
    • course outline
    • exercises / data required
    • sections with potential for 'non-traditional' teaching, HUB style
  • Venue and Date
    • Matt will bring BioQuant seminar room details and schedule.
  • Any other business.



Outline of course as discussed at meeting: File:OutlineCourse v1.txt