Conceptual Question About Miniasm

[chocotwig]

Hi,

I have a conceptual question about miniasm.

Say I have several thousand genomic reads, and I know that 100 of these reads represent a given genomic region, and that in the PAF minimap2 output, the best mappings for each of these 100 reads is ONLY to one of these other 100 reads, and not to any read outside these 100. (although there may be other mappings for these reads outside the 100, these may not be as good).

Then, if I were to input a PAF file for only these 100 reads into miniasm, would the resulting contig(s) for this 100-read genomic region be the same, as compared to the scenario where I would input the PAF file for all the several thousand reads?

This would help deciding how to pre-filter our reads before inputting them into miniasm. Thanks.