search

yyoshiaki / VIRTUS

A bioinformatics pipeline for viral transcriptome detection and quantification considering splicing.
https://github.com/yyoshiaki/VIRTUS/wiki
Other
34 stars 9 forks source link

Cluster methods used by VIRTUS and heatmap issue #34

Open patrick-douglas opened 1 year ago

patrick-douglas commented 1 year ago

Dear yyoshiaki,

I`m having a issue during a VIRTUS run, I have two cases:

1° case: I have 14 samples divided in 3 groups, Group 01: 4 samples Group 02: 06 samples Group 03: 04 samples

The final heatmap show only 11 samples

image

2° case: I have 12 samples divided in 3 group, Group 01: 04 samples Group 02: 04 samples Group 03: 04 samples

The final heatmap shows only 09 samples

image

How can I fix the heatmap to show all samples?

I also would like to know which clustering method is used by VIRTUS, I've tried some (e.g. Complete linkage, average linkage, etc.), but none reproduced the same heatmap generated by VIRTUS, even using only samples in the original VIRTUS heatmap.

Another question is, should I use the summary.csv for heatmap construction?

Thank you in advance

Patrick.

yyoshiaki commented 1 year ago

Hi Patrick,

Sorry, as I cannot reproduce the problem, I have no idea about it. Can you modify VIRTUS_wrapper.py or, by scratch, make figures? Using summary.csv is a simple solution.