[question] use of max_genome_size?

[Svnipni]

I've had great usage of this pipeline on my isolates. But one isolate sample seems to always return a 11Mbp genome on assembly where a typical P.protegens genome size of ~6.5Mbp is expected. So I tried to run

bactopia --R1 B_1.fastq.gz --R2 B_2.fastq.gz --sample sample --species "Pseudomonas protegens" --genome_size median --max_genome_size 7000000 --min_coverage 100 --datasets datasets --outdir out/

But it still returns the same size genome.
Checkm indicated there's some (5.43%) contamination and both Kmer and Kaiju queries on the fastq.gz reads shows a near full hit (70%) matching P.protegens, with some Citrobacter spp sprinkled in.
It seem max_genome_size doesn't affect the assembly, or perhaps I'm not using it properly?

Any advice would be greatly appreciated.

[rpetit3]

Yo @Svnipni !

This looks like a bug I need to look into, because there should be a check after the assembly to prevent this from happening.

Also, this looks like its on v2, do you mind trying on v3 as well?

Thank you!
Robert

[Svnipni]

Yes, this was still V2. Will try it with V3 on Monday!

[rpetit3]

sounds good, thank you!

[Svnipni]

I'm still getting a 11Mbp genome from what I expected to be no larger than 6-7Mbps.
I ran
bactopia --r1 B_1.fastq.gz --r2 B_2.fastq.gz --sample sample_Pprotegens_v3 --species "Pseudomonas protegens" --genome_size 6000000 --max_genome_size 8000000 --outdir bactopia_out/

I'm pretty certain it's my sample that's causing the problem, since SPADES gives a similar genome.