search

yollct / spycone

Spicing-aware time-course network enricher - exploratory analysis for transcriptomics and/or proteomics time series data
GNU General Public License v3.0
10 stars 0 forks source link

Clustering of total isoform usage missing one time point #7

Open NikoLichi opened 1 month ago

NikoLichi commented 1 month ago

Hi there,

Using the Spycone tutorial and my own data, I noticed that there is one missing time point when plotting the clusters for the total isoform usage. Is this expected? For instance, If you see the tutorial cluster figure https://spycone.readthedocs.io/en/latest/alternative.html you have in total 4 time points in the test set, but when plotting the clusters there are only 3. Any idea how to solve this?

Also, the parameters to change axis labels and x-axis tick labels are not working, so it is completely impossible to see which timepoint is missing. Screenshot 2024-05-22 at 10 24 20

Thanks and all the best, Nicolas

NikoLichi commented 1 month ago

I found an answer to my first question in case someone needs it. Taken from the author's paper: Each cluster is represented by a cluster prototype, which is the median change of total isoform usage per pair of time points.

So if you have 4 time points, the DTU cluster plot would be tp0-tp1, tp1-tp2, tp2-tp3, meaning that there are only 3 ticks in the x-axis.