What are "must have" data types one should extract from WGS .fq data?

Question

[optimistsso4co3]

I have multiple .fq whole genomes prepared for variant calling. However, it is quite expensive to repeat the whole pipeline so i wonder what data types are must have to extract?

Right now i am planning to extract the following:

1. Variants (indels, SNPs), with HaplotypeCaller
2. Structural variants (<1000 bp long), with Manta

What are some other "must have" i could extract?

[Pierre Lindenbaum]

why just <1000 bp long ?

[optimistsso4co3]

It's the maximum length generated by Manta for my dataset (~100 bp reads).

[WouterDeCoster]

Do you have a reference for that?

[optimistsso4co3]

Nope, i manually measured the length of structural variants after conversion to PLINK format. Now when i think about it i see there might be a crack in my logic due to PLINK limits on allele naming length.

[Pierre Lindenbaum]

bcftools query -f '%INFO/SVLEN\n'  in.vcf | tr -d '-' | sort -n | tail