search

saeyslab / CytoNorm

R library to normalize cytometry data
33 stars 6 forks source link

K-mean clustering instead of FlowSOM #19

Closed EAC-T closed 3 years ago

EAC-T commented 3 years ago

Hi everyone, For some reason FlowSOM can't cluster my T cells into naive and memory, K-mean can, is there a way to use K-mean clustering instead? Thank you a lot

tomashhurst commented 3 years ago

Hi @EAC-T ,

Our package Spectre uses CytoNorm as the primary batch alignment method. In an upcoming update, we'll have an update where you can perform alignment with whatever clustering approach you want (K-means, phenograph, etc). It's not exactly the same implementation as CytoNorm, but it's similar. Would this work for you?

Tom

EAC-T commented 3 years ago

Hi @tomashhurst

Yes that would be great!

Thank you

SofieVG commented 3 years ago

Hi @EAC-T ,

I'm afraid that would have been out of scope for our implementation, so I'm very happy to hear @tomashhurst will be able to provide this in the Spectre package!

All the best, Sofie