Forum:Survey: Gene Expression Data Analysis Learning Effort

Question

[fac2003]

I'd like to measure the effort typically involved in teaching beginners basic gene expression data analysis (e.g., RNA-Seq). I have some ideas of course from teaching at our institution, but I am trying to get more quantitative information across the community. I created a mini-survey of four questions. If anybody could answer this survey it would be very helpful.

This is an open-survey and you will see all responses after submitting your answer. Note that you can also answer the survey if you are a trainee who took such a course.

Thanks in advance

[Devon Ryan]

FYI, a heatmap of DE genes (one of the stated goals) is typically not that informative. Inevitably you find that the DE genes differ by color between group...which is what you already knew.

[fac2003]

I think heatmaps are useful to help rule out problems with lack of consistency of genes called DE (remember early methods calling significant genes changed in only one sample of a group?). Heatmap are useful as a diagnostic test to check that the stats are working as you expected them to. I'd rather build a heatmap that does not tell me more than I expected rather than not building a heatmap and failing to recognize the approach did not produce the results I expected on a specific dataset. Of course, this is very basic and there is a lot that can be done, but we are talking about courses for complete beginners.

In any case, if you teach complete beginners data analysis of similar scope (even if you make them generate other visualizations), I'd appreciate your input to the survey. Thanks.