Hi all, We are currently running a 16S metagenomics workshop at my institution (North-West University, South Africa), and I combined various lessons from the Carpentries and modified most of them according to our main (critical) lesson's dataset that we were using for 16S analysis. We are covering the following topics: spreadsheet organization, shell, HPC for 16S analyses (using Shi7, NINJA and QIIME), R genomics and specific plots of 16S data in R.
The link to our website is https://nwu-eresearch.github.io/2017-10-24-ARC-16S/ and the various lessons are linked. We will write a blog post after the workshop, but here some notes (thus far): - We needed to extend the workshop to 3 full days to fit all the lessons and decided on an additional half-day where they have the opportunity to work on their own data, and also allow some extra time if we needed it for the lessons. - I made an R presentation of the spreadsheet organization Ecology lesson of DC to guide one of our newly-trained instructors who was teaching for the 1st time. The .html can be viewed in a browser, thus not needing RStudio to view the presentation. This also cut back a few minutes, which we then used for the shell lesson. - At the last minute, I decided to switch from the extensive SWC shell lesson to the shortened genomics one of DC (due to limited time allocated for this lesson) and only modified the output according to our HPC. I used the lesson as is, which was a blessing, since the instructor deleted some .fastq files to demonstrate the rm command, and our 16S data was safe for the next session. Feedback shows that they still need a little more time on this lesson. - One full day for 16S analyses was perfect - we finished in time even with some troubleshooting along the way. This workflow was written by Tonya Ward (Knights Lab | University of Minnesota) and all the required software was installed on our HPC prior to the workshop. - We just finished the R lesson (R genomics from DC modified according to the metadata/mapping file that we used for the 16S analyses). Not all the data in this mapping file is real - I made up some variables in order to do plotting (this lesson can certainly be improved by adding more variables). Time allocation seemed to be perfect, the pace was not fast at all and almost everybody kept up with the instructor. - In the next session, they will use output files generated by QIIME to make a variety of visualizations in R. Another newly-trained instructor will be teaching this, and he wrote the workflow for us. Any comments/suggestions/questions are welcome! I hope this is helpful in some way. A detailed blog post will follow in the next week or two. Kind regards, Bianca On Mon, Aug 22, 2016 at 9:45 AM, Carlos Martinez Ortiz < [email protected]> wrote: > Hi Jon, > > > The data carpentry ecology lessons do try to use the same (or very > similar) data. So doing OpenRefine -> SQL -> SQL from Python, works very > well. In my opinion, when you are able to run these lessons one after the > other, it makes the workshop more coherent for the learners, but the > lessons are still complete and can stand on their own. > > > Cheers, > > Carlos > > > ------------------------------ > *From:* Discuss <[email protected]> on behalf > of Jon Pipitone <[email protected]> > *Sent:* 20 August 2016 16:56:43 > *To:* [email protected] > *Subject:* [Discuss] Teaching lessons around a single example > dataset/scenario > > Hi all, > > A few of us at the Centre for Addiction and Mental Health in Toronto > have been teaching a series of Software Carpentry-like workshops[1] > (actually some of them are exactly SWC workshops) over two weeks aimed > more specifically at researchers in our organization. Much of what we > teach is very introductory: what is programming, how to use the linux > shell, a very basic intro to R focusing on statistics, a MATLAB and SPSS > primer, using Photoshop, etc.. We do also get to a few more advanced > topics in some workshops: e.g. doing fMRI analysis in python, using a > compute cluster. > > We received feedback from learners and instructors that having more > cohesion between the lessons would be really helpful to tie things > together (currently we have a mixture of lessons with toy examples, more > elaborate worked examples, and some with only descriptions/powerpoint), > but it's disjointed: there isn't a theme or example dataset running > through the workshops. > > Has anyone tried creating lessons for several different topics around a > single example scenario? E.g. using Nelle's data from the Shell > lectures[2] to also teach R, Python, Git, etc.. How has it worked out? > Is there anything we should be wary of as we wander down this road? > > Thanks! > Jon. > > [1] e.g. https://camh-scwg.github.io/compucool > [2] http://swcarpentry.github.io/shell-novice/01-intro/ > _______________________________________________ > Discuss mailing list > [email protected] > http://lists.software-carpentry.org/listinfo/discuss > > _______________________________________________ > Discuss mailing list > [email protected] > http://lists.software-carpentry.org/listinfo/discuss >
_______________________________________________ Discuss mailing list [email protected] http://lists.software-carpentry.org/listinfo/discuss
