Hi Ben,

I've been competitively mapping reads to a collection of ~7,000 draft genomes using bwa-mem2. Recently, I've tried switching to coverm to complete both mapping (still with bwa-mem) and depth calculation (seqdepth calculation is much faster than the ad hoc python scripts I was using previously) for these genomes.

My longterm aim is to run multiple metagenomes through coverm but since my genome set is so large, I want to avoid building a new index each time I start the process over with a new metagenome. So I wanted to offer coverm genome my bwa index via -r for each run. Based on reading the help documentation, I'm attempting to call coverm genome in this way:

coverm genome -1 ${fwd} -2 ${rev} -p bwa-mem -r ${index} -d ${g_dir} \
--min-read-percent-identity 95 --min-read-aligned-percent 75 --output-format sparse \
--trim-max 90 --trim-min 10 -m trimmed_mean -o ${output} -t 24

Where the variables follow:
fwd/rev: forward and reverse gzipped FASTQ,
index: path to a bwa-mem index, including the stem
g_dir: path to directory containing genomes (suffix matches default .fna)
output: what I want the output .tsv saved as

When I run this, coverm is not able to recognize my BWA index (passed with my index variable above) despite a lot of fiddling on my end. I see the following error before coverm exits:

thread 'main' panicked at 'The reference specified '/storage/coda1/p-ktk3/0/shared/rich_project_bio-konstantinidis/shared3/mst/01_pipeline/xx_references/xx_fulldbs/MST_db' does not appear to exist', src/mapping_parameters.rs:159:9
note: run with `RUST_BACKTRACE=1` environment variable to display a backtrace

Any thoughts or feedback would be greatly appreciated.

Cheers,

Blake