I am doing a metagenomic data analysis.

Where from cell free DNA of AML patients who have sepsis.

It is a illumina NOVA Seq paired end data.

When I used various algorithms like minimap2, bowtie2 etc I got mapped reads of each TAXA.

What's the best way to predict Species abundance from its number of mapped reads ?

I have ref seq length of species also.

Some do say abundance = mapped reads/ ref seq length.

I wanted to know if there is any literature of how abundance could be predicted which is a more dynamic and robust quantitative value ?

happy to engage !