Question about why the cluster number is set to 22 when clustering the mouse embryo dataset

JinmiaoChenLab / GraphST

GNU Affero General Public License v3.0

84 stars 8 forks source link

Question about why the cluster number is set to 22 when clustering the mouse embryo dataset #24

Open tqwang743 opened 5 months ago

tqwang743 commented 5 months ago

Hi,

It is a fantastic method for analyzing spatial transcriptomics!

In paper, you say "Although the original annotation had 12 reference clusters (Fig. 3A), we set the number of clusters in our testing to 22 to acquire a higher resolution of tissue segmentation.". I wonder if there are other reasons for setting cluster number to 22 besides getting a higher resolution, or if there have been comparison experiments that set it to a value other than 22.

Thank you very much!

longyahui commented 4 months ago

Hi,

Thank you for your question.

For mouse embryo dataset, we found increasing cluster number will match the marker genes better. Hope it helps.

Best regards,
Yahui