Metaphlan flags and parameters for extreme metagenomes

Microbial community profiling MetaPhlAn
1

Hey, I do apply metaphlan on my desert microbiomes yet the resulting profiles are extremely y low in quality, my default flags was

    --input_type fastq \\

    --nproc "$SLURM_CPUS_PER_TASK" \\

    --bt2_ps very-sensitive \\

    --bowtie2out "${SAMPLE}.bowtie2.bz2" \\

yet now considering to add

metaphlan “${R1},${R2}” \

    --stat_q 0.05 \\

    --perc_nonzero 0.01 \\

as well, do you think these will be problematic, since I deal with extreme bacteria I need loosen parameters to have a better range of data in the end? much appreciated

2

Your desert microbiome likely differs substantially from the references used by metaphlan, which are biased towards human and gut microbiomes.
You could try reducing the stringency of the thresholds, however, it (1) may not improve the results and/or (2) may lead to erroneous results.

1 Like