IQ-Tree error when running 04-mask_pseudogenome.sh

[bsalehe]

Hi @tavareshugo and @avantonder. I am continuing to test the pipeline/scripts as participant with M_tuberculosis isolate. I am running script 05-run_iqtree.sh. Unfortunately. I keep getting this weird error:-

terminate called after throwing an instance of 'std::cxx11::basic_string<char, std::char_traits, std::allocator >' ERROR: STACK TRACE FOR DEBUGGING: ERROR: 1 funcAbort() ERROR: 2 () ERROR: 3 gsignal() ERROR: 4 abort() ERROR: 5 gnu_cxx::verbose_terminate_handler() ERROR: 6 () ERROR: 7 () ERROR: 8 cxa_rethrow() ERROR: 9 () ERROR: 10 Alignment::addConstPatterns(char) ERROR: 11 runPhyloAnalysis(Params&, Checkpoint, IQTree&, Alignment&) ERROR: 12 runPhyloAnalysis(Params&, Checkpoint*) ERROR: 13 main() ERROR: 14 __libc_start_main() ERROR: 15 () ERROR: ERROR: IQ-TREE CRASHES WITH SIGNAL ABORTED ERROR: For bug report please send to developers: ERROR: Log file: results/iqtree/Nam_TB.log ERROR: Alignment files (if possible) scripts/05-run_iqtree.sh: line 26: 68670 Aborted (core dumped) iqtree -fconst results/snp-sites/constant_sites.txt -s results/snp-sites/aligned_pseudogenomes_masked_snps.fas --prefix results/iqtree/Nam_TB -nt AUTO -ntmax 8 -mem 8G -m GTR+F+I -bb 1000

I wonder if any you experienced this error. Any idea what is wrong with inputs or command. Below is the script I modified after successfully running the 04-mask_pseudogenome.sh script.

#!/bin/bash

#mamba deactivate
#mamba actiavte iqtree

# create output directory
mkdir -p results/snp-sites/
mkdir -p results/iqtree/

# extract variable sites
snp-sites results/bactmap/masked_alignment/aligned_pseudogenomes_masked.fas > results/snp-sites/aligned_pseudogenomes_masked_snps.fas

# count invariant sites
snp-sites -C results/bactmap/masked_alignment/aligned_pseudogenomes_masked.fas > results/snp-sites/constant_sites.txt

# FIX!!
# Run iqtree
iqtree \
  -fconst results/snp-sites/constant_sites.txt \
  -s results/snp-sites/aligned_pseudogenomes_masked_snps.fas \
  --prefix results/iqtree/Nam_TB \
  -nt AUTO \
  -ntmax 8 \
  -mem 8G \
  -m GTR+F+I \
  -bb 1000

Cheers!

[avantonder]

Two thoughts:

Is there anything in the constant_sites.txt file?

Is there sufficient memory on the machines?

[bsalehe]

Two thoughts:

Is there anything in the constant_sites.txt file?

Is there sufficient memory on the machines?

Yes, with cat results/snp-sites/constant_sites.txt I get the following output

692542,1311546,1307543,691941

[avantonder]

Hmm, does it at least read in the alignment before crashing?

[tavareshugo]

It could be a memory issue. We're currently running a course in the room, so these machines are potentially quite stretched. You could try running free -h to see how much free memory there is at the moment. For example, on my test instance I only have 7G free at the moment.

[avantonder]

@bsalehe The potential cause of the problem occurred to me on my drive home. The -fconst flag expects the contents of constant_sites.txt not the file so two ways to edit the script:

-fconst 692542,1311546,1307543,691941 (copy and paste contents of constant_sites.txt)

-fconst $(cat results/snp-sites/constant_sites.txt) (directly input contents of constant_sites.txt)

[bsalehe]

This -fconst $(cat results/snp-sites/constant_sites.txt) did the job. So, do we need to update the script for participants?

Thanks both!

[avantonder]

In the answer to the exercise I've used -fconst $(cat results/snp-sites/constant_sites.txt). Perhaps I should change the example code above in the course materials so it's not 692240,1310839,1306835,691662

[tavareshugo]

I think using the numbers might be a bit easier. As I revise the materials, I can adjust the exercise to be more explicit that they are supposed to copy the contents of the file into the command, not use the path to the file.

[tavareshugo]

Humm... I think we will need to include $(cat results/snp-sites/constant_sites.txt) after all, because the script itself generates the file, so there's no in-between step where they calculate these values and look at them.

Maybe instead I can include a couple of FIX_INPUT_FASTA_FILE for the snp-sites input files. So, at least they can think about what should be used as input at these steps.

[bsalehe]

Humm... I think we will need to include $(cat results/snp-sites/constant_sites.txt) after all, because the script itself generates the file, so there's no in-between step where they calculate these values and look at them.

Maybe instead I can include a couple of FIX_INPUT_FASTA_FILE for the snp-sites input files. So, at least they can think about what should be used as input at these steps.

I included $(cat results/snp-sites/constant_sites.txt) when I was running script for S_aureus. I agree.