biobakery / phylophlan

Precise phylogenetic analysis of microbial isolates and genomes from metagenomes
https://huttenhower.sph.harvard.edu/phylophlan
MIT License
128 stars 33 forks source link

Bootstrap error obtained. zsh: killed #97

Open pavlo888 opened 2 years ago

pavlo888 commented 2 years ago

Hi @fasnicar

I have been trying to calculate the bootstrap for a phylogeny of ~200 bacterial genomes using Phylophlan and following the recommendations from a previous post (https://github.com/biobakery/phylophlan/issues/70). However, I cannot manage to obtain any results. So far, I have tried it two times using the following commands:

raxmlHPC-PTHREADS-SSE3 -s Agro_tax_217_fna_concatenated.aln -n tree_w_bootstraps -f a -m GTRCAT -# 100 -p 1989 -x 1989 T 8 raxmlHPC-PTHREADS-SSE3 -s Agro_tax_217_fna_concatenated.aln -n tree_w_bootstraps2 -f a -m GTRCAT -# 100 -p 1989 -x 1989 T 4

I eventually get the following error message:

zsh: killed raxmlHPC-PTHREADS-SSE3 -s Agro_tax_217_fna_concatenated.aln -n -f a -m GTRCA zsh: killed raxmlHPC-PTHREADS-SSE3 -s Agro_tax_217_fna_concatenated.aln -n -f a -m GTRCA

When I ran the command the first time, the procedure finished all 100 bootstrapping and was on the Fast ML tree optimisation. The second time I ran the command, the procedure was on the 51st bootstrapping.

Do you have any idea what the cause of the error might be and if I can fix it? I am running the package in a laptop MacBook Air (processor: 1,6 GHz Dual-Core Intel Core i5, and memory: 8 GB 1600 MHz DDR3) and I am using an external disk for storing the output files.

Thank you for your help!

Cheers, Pablo

fasnicar commented 2 years ago

Dear Pablo,

In my experience, when the bash returns a message with killed it usually means that there are not enough resources and the OS decided to kill a job that takes a large fraction of them. So, I would suggest you verify that the memory is enough in this case. You can also use the RAxML webpage https://cme.h-its.org/exelixis/web/software/raxml/ to estimate the memory requirements in your case.

Many thanks, Francesco