ablab / quast

Genome assembly evaluation tool
http://quast.sf.net
Other
395 stars 76 forks source link

Error while creating circos plot #114

Open ghost opened 4 years ago

ghost commented 4 years ago

Dear developers,

After the following step:

Creating large visual summaries...
This may take a while: press Ctrl-C to skip this step..
1 of 3: Creating Icarus viewers...
2 of 3: Creating Circos plot...

Quast throws an error and exits:

list index out of range
Traceback (most recent call last):
File "/opt/quast-5.0.2/quast.py", line 307, in <module>
return_code = main(sys.argv[1:])
File "/opt/quast-5.0.2/quast.py", line 265, in main
features_containers, cov_fpath, os.path.join(output_dirpath, 'circos'), logger)
File "/opt/quast-5.0.2/quast_libs/circos.py", line 560, in do
conf_fpath, circos_legend_fpath = create_conf(ref_fpath, contigs_fpaths, contig_report_fpath_pattern, output_dir, gc_fpath, features_containers, cov_fpath, logger)
File "/opt/quast-5.0.2/quast_libs/circos.py", line 443, in create_conf
cov_data_fpath, cov_points = create_coverage_plot(cov_fpath, window_size, chr_lengths, data_dir)
File "/opt/quast-5.0.2/quast_libs/circos.py", line 257, in create_coverage_plot
cov_by_chrom[chrom][pos // window_size].append(depth)
IndexError: list index out of range
ERROR! exception caught!

I can not understand the reason for this. How can I overcome this? Thank you in advance. Kind regards, Marios

alexeigurevich commented 4 years ago

Dear Marios, The problem seems rather strange but maybe it is a bug in our code. Most likely it is somehow related to your reads and their alignment to reference genome.

However, to reproduce and fix the issue we need sample data from your side. Could you attached them here (direct upload or via a link to some other host place) or send us privately via quast.support@cab.spbu.ru?

Ideally, it should be a minimal example that still results in the error message. E.g. full reference and first 1K or 10K reads. For downsampling, you can use, e.g. seqtk sample. However, we can work with the full dataset as well if it is more suitable for you.

ghost commented 4 years ago

This is the way we run quast:

python /opt/quast-5.0.2/quast.py \
-o /home/mgabriel/Downloads/data/new_refDV2/output \
-r /home/mgabriel/Downloads/data/new_refDV2/GCF_000005245.1_dvir_caf1_genomic.fna \
--features /home/mgabriel/Downloads/data/new_refDV2/GCF_000005245.1_dvir_caf1_genomic.gff \
--threads 16 \
--eukaryote \
--large \
--circos \
--conserved-genes-finding \
--gene-finding \
--rna-finding \
--fragmented \
--pe1 /home/mgabriel/Downloads/data/DroVir2/SRR1536175_1.fastq.gz \
--pe2 /home/mgabriel/Downloads/data/DroVir2/SRR1536175_2.fastq.gz \
/home/mgabriel/Downloads/data/DroVir2/flye/assembly.fasta

GCF_000005245.1_dvir_caf1_genomic.fna and GCF_000005245.1_dvir_caf1_genomic.gff were downloaded from ftp://ftp.ncbi.nlm.nih.gov/genomes/all/GCF/000/005/245/GCF_000005245.1_dvir_caf1

SRR1536175_1.fastq.gz and SRR1536175_2.fastq.gz were downloaded from https://www.ebi.ac.uk/ena/data/view/SRR1536175

assembly.fasta was produced using MaSuRCA. The format is:

>contig_1
ATGTTTATAATTGTGCTTATGCTTATGTGCTAAGTTATGTTAAAGGGTGGGTGCGACTAT
GGATATGTCACTCCCCCAACCATTAGAAAAGTGCCTGTCCTCAGGTGTTTAAGTTTGGAT
...
>contig_2
TTTGGTCAAGTAACAGCAGAGATAACACCGACGGCTGCGGACATCATCCAACAGCAGAAA
ATAACAATCATAATTACAGCCTAAGAGCAAATAGTTTTTTTGCGCGCTCTTTGTAAGCTG
...

I would also like to add that after this error message I installed circos independently from quast according to the instructions of circos (http://circos.ca/software/installation/). Now circos is in my path and I can execute it. Thank you for your quick response.

bersbers commented 4 years ago

Hello, I am experiencing exactly the same problem, could you tell me what was the cause and how did you solved please?. I do have circos running independently and I am sure QUAST can access it. THANK YOU!

ghost commented 4 years ago

I just installed circos (http://circos.ca/software/installation/) and then I tested if it was installed correctly. Finally, I added circos in my path and QUAST was able to run successfully.

bersbers commented 4 years ago

Thanks, CIRCOS is def. in my path but still ... I will keep trying ...