Hi,
I have been trying to use SPATA2 on simulated data. The method I am using generates data with location coordinates in the interval of 0-1. When run with these coordinates SPATA2 identifies all genes as significant in the annotation gradient screening. I thought this could be due to the small resulting distances so I tried easy scaling using the coordninated of SPATA2 example data. This way SPATA2 identifies only 2/310 genes as significant. When I divided the scale factor by 2, all genes are identified again. As I understand it then, the gene identification strongly depends on the ditances between cells/spots. I was wondering if you could help me with finding a plausible scale factor for this.
Hi,
I have been trying to use SPATA2 on simulated data. The method I am using generates data with location coordinates in the interval of 0-1. When run with these coordinates SPATA2 identifies all genes as significant in the annotation gradient screening. I thought this could be due to the small resulting distances so I tried easy scaling using the coordninated of SPATA2 example data. This way SPATA2 identifies only 2/310 genes as significant. When I divided the scale factor by 2, all genes are identified again. As I understand it then, the gene identification strongly depends on the ditances between cells/spots. I was wondering if you could help me with finding a plausible scale factor for this.