miicTeam / miic_R_package

Learning causal or non-causal graphical models using information theory
GNU General Public License v3.0
26 stars 3 forks source link

Are there descriptions to the lines connecting the nodes? #126

Closed KevinZhou-hub closed 1 year ago

KevinZhou-hub commented 1 year ago

Hi! Thank you for the great package! I'm running MIIC on the miic server, the results generated only had labels on the nodes, but had no description to the lines connecting the nodes. I read in some other publications had descriptions like this: 2023-02-21 112903

2023-02-21 112845 I looked into your tutorial and user guide and none of them seem to have described what the lines and arrows mean. How should I label my lines? Where can I find the descriptions to my lines? Any suggestion would be helpful! Thank you!

louise-rb-dupuis commented 1 year ago

Hi @KevinZhou-hub thank you for your question ! We will come back to you first thing next week but in the meantime maybe you can have a look at our publications at http://kinefold.curie.fr/isambertlab/publications.htm for the method part if you want !

KevinZhou-hub commented 1 year ago

Okay!Thank you for the quick reply!

louise-rb-dupuis commented 1 year ago

Dear @KevinZhou-hub , sorry for the delayed response , we wanted to first put our latest paper on arxiv: https://arxiv.org/abs/2303.06423 as we hope it will give you a better presentation of the new features of MIIC (for instance the genuine "green" vs putative "blue/red" arrowheads, see Box1).

Here is a summary of the links you can find in MIIC networks : MIIC_arrowheads

Finally the color blue or red results from the Pearson correlation calculation between X and Y to highlight if the identified effect is positive (corr >0) or negative (corr <0). The link will be grey in case of categorical variables with levels that are not ordered in the state order file.

I hope this answer everything, don't hesitate to reach again !