cellrouter icon indicating copy to clipboard operation
cellrouter copied to clipboard

pretty low GRN scores obtained

Open JianmeiZhong opened this issue 6 years ago • 14 comments

Hi, I work through most part of the analysis pipeline but the results of GRN scores obtained via grnscores() in specified paths SP_10.SP_8 as tSNE and barplot shows.the grn scores have different levels compared to yours presented in several exemplary analysis,that's level of 0.0* versus 0.* .Sometimes,the GRNscores plot has one or two genes with low scores in both directions.I don't know if there's some wrong configuration in my pipeline.If my source and endpoints are correct and reliable,low scores level are possible for some datasets and it's available? barplot another questions asking for your guide :) known knowledge this project based on are less subpopulations and increasing number of cells along differentiation process.So,k=120 was set according to my understanding in function buildGRN() and creatKNN() to obtain relatively less subpopulations as tSNE plot shows.k value is different a lot from yours.It might be a cause of low GRN scores?And if a path didn't go through every subpopulation in non-branched cells differentiation,gene expression dynamics with this path can represent entire process of changes? tsne

Thank a lot !

Jm Zhong

JianmeiZhong avatar Apr 20 '18 13:04 JianmeiZhong