ivar_variants_to_vcf.py collapses low quality ivar variant with high quality variant

[Rohit-Satyam]

Description of the bug

I noticed that ivar_variants_to_vcf.py is not behaving correctly. I was testing the difference between the phasing features of your ivar_variants_to_vcf.py and phasing.py, provided by COVIGATOR and it turns out that ivar_variants_to_vcf.py collapses bad quality variant with good quality variant
Eg:

MN908947.3	24129	A	G	30	29	36	1	1	37	0.0322581	31	0.508197	FALSE	CDS:ENSSASP00005000004	AAC	N	AGC	S
MN908947.3	24130	C	A	0	0	0	33	32	36	1	33	2.7703E-19	TRUE	CDS:ENSSASP00005000004	AAC	N	AAA	K

The A>G mutation at position 24129 is present in just 1. Therefore no matter that low quality variant call lie within same codon as high quality, they should be kept separate and only the high-quality calls must be collapsed. The current behaviour of your script is as follows:

MN908947.3	24129	.	AC	GA	.	ft	DP=31	GT:REF_DP:REF_RV:REF_QUAL:ALT_DP:ALT_RV:ALT_QUAL:ALT_FREQ	1:30:29:36:1:1:37:0.0322581	
MN908947.3	24256	.	T	C	.	ft	DP=33	GT:REF_DP:REF_RV:REF_QUAL:ALT_DP:ALT_RV:ALT_QUAL:ALT_FREQ	1:32:0:36:1:0:37:0.030303	

However it should be

MN908947.3	24129	.	A	G
MN908947.3	24130	.	C	A

Also, it is quite misleading to assign AC>GA with following tags of high quality variant DP=31 GT:REF_DP:REF_RV:REF_QUAL:ALT_DP:ALT_RV:ALT_QUAL:ALT_FREQ 1:30:29:36:1:1:37:0.0322581

Command used and terminal output

No response

Relevant files

No response

System information

No response

[Rohit-Satyam]

@mattheww95 @saramonzon Any thoughts here?

[drpatelh]

@Rohit-Satyam presumably this issue is unrelated to #321 (comment) ? @saramonzon is this something we can fix in future releases?

[saramonzon]

Yes this is unrelated to the #321 fixes, we can totally fix it in the next releases, i will add it to the to do list :) Thanks for the tip!

[drpatelh]

Cool. Will move to the next milestone. Thanks @saramonzon !