emmahodcroft / cluster-picker-and-cluster-matcher Goto Github PK

Hi, @emmahodcroft @hxnx-sam

When I was trying to convert to the MCC tree to newick in R, I met an error like this.
The code is from here:

Warning message in matrix(tips, 2, length(tips)/2):
"data length [1027] is not a sub-multiple or multiple of the number of rows [2]"
Error in jj[[1]]: subscript out of bounds
Traceback:
1. MCC_to_NWK(mccName)

Do you know how to solve it?
Yan

In the GUI and in interactive command-line (run with no arguments), genetic distance needs to be supplied as a %, ex: 4.5, and is divided by 100 (see line 984 of ClusterPicker.java and 126 of ClusterPickerGUI.java).

However if provided with arguments in command-line mode, genetic distance is NOT divided by 100 (see line 933 in ClusterPicker.java) and so needs to be provided not as a % (ex: 0.045)

This is confusing and inconsistent, and we should probably try to rectify this - but without messing up backwards compatibility too much!

I'm not sure how time-consuming this step is, so not sure if it is negligible:

https://github.com/emmahodcroft/cluster-picker-and-cluster-matcher/blob/master/src/clusterPicking/ClusterPicker.java#L456

But you are storing the things you want to process in an ArrayList, but you're iteratively removing from the front of the ArrayList, which is O(n) each time (where n is the number of elements in the ArrayList). You should use a Queue (which is implemented as a LinkedList) for O(1) removal from the front, given that you don't care about random access of elements

Thanks for providing this very useful tool! However, reporting of genetic distances in clusters are in the format of 0.0 and is not very useful when analyzing SARS-CoV-2 diversity (with a "genetic distance threshold" of 0.05%) - is it possible to adjust so more digits are reported back?
I'm using version 1.2.1
Thank you!
Cheers
Patrik Medstrand, Lund University, Sweden

It would be great if you could make a release on the "Releases" tab in Github.

This will help us to package it in Conda and Homebrew.

Thank you

Hi

Samantha Lycett (Cluster Picker): [email protected]
Manon Ragonnet: [email protected]
in the user manual are no longer available

Thanks for the ClusterPicker software. We are using it as part of our Hepatitis A Virus infection cluster detection pipeline (https://github.com/schultzm/HAVIC).

In using it, we are starting to notice the finer details. One of which we are uncertain about. If I set the genetic distance measure as 3/300, so input to the command line as 0.01, will clusterpicker place isolates with for example 0.013 divergence in the same cluster? We have a case where two isolates are 5/367 apart, but in a clade at 100% support. So, these two isolates get clustered into the same cluster, even though this translates to >0.01 divergence. My question then is, if I input the cutoff to cluster picker as a float with four significant digits, say 0.0100, will clusterpicker then recognise that 0.0136239782 (5/367) is in fact greater than 0.0100?

Thanks,