cannot run orthofinder when activating R4.3 environment using conda

[Angelia111]

Hi JT Lovell，

I am tring to learn how to use GENESPACE recently, so I start from the test run. Everything is OK until I run the command " gpar <- init_genespace(wd = wd,path2mcscanx = path2mcscanx)". There is no error reported, but I cannot get the directroy orthofinder under working directory wd.

I know this should be the problem with orthofinder. However, it is already installed under conda environment 'orthofinder' using your codes. Every time when I activate this environment, orthofinder is working. I have also added the full path into .bashrc file. But if I didn't activate it first, it cannot work, and with a problem of module 'scipy'. But I checked scipy is installed.

The version of my R under conda/bin is 3.6.3， when I was trying to install Biostrings dependency, I found this R version is not high enough. So I create a new conda environment R4.3 to install R 4.3.1. Everytime when I open R4.3.1, I need to open this environment first. Therefore I guess this is the problem, because we cannot open two different environments, right or no?

I also tried to install orthofinder under R4.3 environment using conda, but I only got a older version 2.4.0. As you can expect, error appears when I run " gpar <- init_genespace(wd = wd,path2mcscanx = path2mcscanx)".

Could you help me with this issue, any help is highly appreciated.

Best regards,

Li

[jtlovell]

Were you able to run the tutorial data? If so, please report the specific error that is reported from that.

[Angelia111]

Hi, Jtlovell,

there is no error, but the orthofinder dependency is not found when I ran this step. However I have already installed this package. I think the main problem is that my orthofinder and my R4.3 were installed under different environments, as I descirbed above. But I have no idea how to solve it?

gpar <- init_genespace(wd = wd,path2mcscanx = path2mcscanx) Checking Working Directory ... PASS: /raid4/yangli/biosoft/genespace_test Checking user-defined parameters ... Genome IDs & ploidy ... chicken: 1 human : 1 mouse : 1 Outgroup ... NONE n. parallel processes ... 16 collinear block size ... 5 collinear block search radius ... 25 n gaps in collinear block ... 5 synteny buffer size... 100 only orthogroups hits as anchors ... TRUE n secondary hits ... 0 Checking annotation files (.bed and peptide .fa): chicken: 18093 / 18093 geneIDs exactly match (PASS) human : 20678 / 20678 geneIDs exactly match (PASS) mouse : 22858 / 22858 geneIDs exactly match (PASS) Checking dependencies ... Found valid path to DIAMOND2 v2.011: diamond Found valid MCScanX_h executable: /raid4/yangli/biosoft/MCScanX/MCScanX_h

[jtlovell]

You can provide a valid path to orthofinder via init_genespace(..., path2orthofinder = 'path/to/program'). But this can be problematic since orthofinder bundles dependencies. I would suggest just putting orthofinder into the path in the env you open R with. John