CAFE5 running error: Invalid lambda index set for 1

134 views
Skip to first unread message

Pu Shi

unread,
Feb 20, 2023, 2:42:36 AM2/20/23
to hahnlab-cafe
Hello,
When I run the CAFE5 with the code: "-i cafe.input.tsv -t 14aatree.fa -y 14aa_lambda.fa -k 3 -o report_14aa/" , the "Invalid lambda index set for 1" will occur. And when I use the speice name like "speice:1" , the error would occur with the "Invalid lambda index set for
 speice". 
After that, I let the CAFE5 calculate the lambda with code " cafe5 -i cafe.input.tsv -t 14aatree.fa -k 3 -l 0.01 -o report_14aa/" , but the error "Failed to initialize any reasonable values" came out. So I'd like to know where are the mistakes in my code.

The cafe.input.tsv, 14aatree.fa, 14aa_lambda.fa, 14aa_lambda_speicename.fa , and the error are  in the attachment. 
Looking forward your reply.
Best Wishes!

14aa_lambda_speicename.fa
error.txt
14aatree.fa
cafe.input.tsv
14aa_lambda.fa

Hahn, Matthew

unread,
Feb 20, 2023, 10:01:35 AM2/20/23
to Pu Shi, hahnlab-cafe
Hi,

It seems that your tree is quite long, and your families are highly variable (with some not likely existing in the root). So I would suggest using only a subset of species and families.


Matt

--
You received this message because you are subscribed to the Google Groups "hahnlab-cafe" group.
To unsubscribe from this group and stop receiving emails from it, send an email to hahnlabcafe...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/hahnlabcafe/cb8948e2-5e2a-4edb-9f82-5228834ea00fn%40googlegroups.com.
<14aa_lambda_speicename.fa><error.txt><14aatree.fa><cafe.input.tsv><14aa_lambda.fa>

Pu Shi

unread,
Feb 20, 2023, 10:18:39 AM2/20/23
to hahnlab-cafe
Hi,
Thanks for your reply, and I just used 14 speices to run the CAFE5, it seems like fine in other researchers CAFE5, and I had removed the family with more than 100 orthologs, the error came the same.
So whether is it better to use the CAFE4?
best wishes!

Hahn, Matthew

unread,
Feb 20, 2023, 11:41:12 AM2/20/23
to Pu Shi, hahnlab-cafe
Sorry, but it’s not the number of species, but how distantly related they are. Likewise, it’s not the size of families, but the variance in copy-number that is the issue.


matt

Reply all
Reply to author
Forward
0 new messages