search

diyabc / diyabcGUI

User interface to DIYABC/AbcRanger
https://diyabc.github.io/
Other
4 stars 2 forks source link

simulated dat #109

Closed tangtang5299 closed 3 years ago

tangtang5299 commented 3 years ago

hello, After the simulation of my SNP data, PCA analysis always showed bad results, as shown in the figure below. Is there any way to improve it? I am looking forward to your reply.

modelchoice_out_graph_lda

gdurif commented 3 years ago

Hi, are you sure the historical models you are considering are consistent with you data ?

tangtang5299 commented 3 years ago

Thank you for your reply，I am not sure. My historical model is three simple models divided into groups according to structure. headerRF.txt

gdurif commented 3 years ago

In this case, it means that either your model, or the priors that you defined are not consistent with your data. You could try to modify either the models or the priors to simulate a more realistic training set regarding your data.

You can also check the file pcaloc1_locate.txt to compare the value of the summary statistics on your dataset and on the training set.

tangtang5299 commented 3 years ago

Thank you. It is appreciated for answering my question in your busy time.