SeqSero2 1.2.1 crashes on some readsets

[evolarjun]

We've noticed SeqSero2 reproducibly crashing in on some readsets such as ERR6806453.

Example:

$ srr=ERR6806453
$ fastq-dump --split-files --defline-seq '@$ac.$si/$ri' --defline-qual + $srr
Read 1520076 spots for ERR6806453
Written 1520076 spots for ERR6806453
$ SeqSero2_package.py -t 2 -m a -p 10 -d ${srr}_out -i ${srr}_1.fastq ${srr}_2.fastq
building database...
mapping...
check samtools version: 1.12
[bam_sort_core] merging from 0 files and 10 in-memory blocks...
assembling...
blasting...

Traceback (most recent call last):
  File "/home/aprasad/miniconda3/envs/seqsero121/bin/SeqSero2_package.py", line 1648, in <module>
    main()
  File "/home/aprasad/miniconda3/envs/seqsero121/bin/SeqSero2_package.py", line 1404, in main
    O_choice,fliC_choice,fljB_choice,special_gene_list,contamination_O,contamination_H,Otypes_uniq,H1_cont_stat_list,H2_cont_stat_list=predict_O_and_H_types(Final_list,Final_list_passed,new_fasta) #predict O, fliC and fljB
  File "/home/aprasad/miniconda3/envs/seqsero121/bin/SeqSero2_package.py", line 846, in predict_O_and_H_types
    O_choice,O_nodes,special_gene_list,O_nodes_roles,contamination_O,Otypes_uniq=decide_O_type_and_get_special_genes(Final_list,Final_list_passed)#decide the O antigen type and also return special-gene-list for further identification
  File "/home/aprasad/miniconda3/envs/seqsero121/bin/SeqSero2_package.py", line 718, in decide_O_type_and_get_special_genes
    highest_O_coverage=max([float(x[0].split("_cov_")[-1].split("_")[0]) for x in final_O if "O-1,3,19_not_in_3,10" not in x[0]])
ValueError: max() arg is an empty sequence