RE: [yuchaojiang/Canopy] Clustering with 1 Sample (#8)

[Jiang, Yuchao]

Hi Jesse,

 

I just replied to another user with the same question yesterday. Please see below. Note that you need to update the package from GitHub. Canopy is not primarily designed for one sample  -- I’m not sure how much information about phylogeny you can get from one heterogeneous sample. Also Canopy is not focused on clustering, which is only a by-product of the program.

 

Thanks,

Yuchao

 

From: Jesse Eaton [mailto:notifi...@github.com]
Sent: Thursday, April 27, 2017 10:49 AM
To: yuchaojiang/Canopy <Can...@noreply.github.com>
Cc: Subscribed <subsc...@noreply.github.com>
Subject: [yuchaojiang/Canopy] Clustering with 1 Sample (#8)

 

I have a single sample around 325 single nucleotide alterations (SNAs) which I've stored in the R and X variables. I can use these with my CNA data to run canopy.sample() but when I run clustering in the following way

canopy.cluster(R = R,

               X = X,

               num_cluster = num_cluster,

               num_run = num_run)

I get the error

Error in while (diff > 0.001 || numiters <= 30) { : 

  missing value where TRUE/FALSE needed

Any reason why this might be? Thank you for your help!

—
You are receiving this because you are subscribed to this thread.
Reply to this email directly, view it on GitHub, or mute the thread.

[Lorraine Soudade]

Hello Yuchao,

I already installed the last version of canopy but I still get this error when I run the canopy.cluster function with my R and my X :

"

Error in while (diff > 0.001 || numiters <= 30) { : 

  missing value where TRUE/FALSE needed

"

If I look at toy3$R, it's the same format that my R matrix except I have two samples. I tried to run this function on your other dataset MDA231, but I get the same error. Is it normal ? How can I fix it ?

Thank you,

[Gene Urrutia]

Hi Lorraine,

Thank you for your interest in Canopy and the Marathon pipeline.

I would suggest lowering the number of clusters.  For the MDA231 data, the function will perform normally if num_cluster is set to 2 instead of 2:9.  There is not enough variability in the data to define many clusters.

If this does not fix your issue, would you send me your image file using save.image(file=‘debug.rda’) along with your code.  Then I will be able to troubleshoot.

Please also note that Canopy is not focused on clustering, which is only a side feature of the program.

Best,
Gene