MAG abundance estimation

At this point we still have no information about how much of each MAG was present in the original samples—this is something we will try to do in this section. To estimate MAG abundance, we will take the original reads and map them back to the recovered MAGs. After that we can count all the reads that mapped to each genome, normalize the counts to genome lengths and present the results using one of the available metrics (RPKM or TPM, more below).

MAG indexing¶

Similarly to how it was done for MAG recovery, we first need to index our dereplicated MAGs. We can do it using the index-derep-mags action from the q2-annotate plugin:

mosh assembly index-derep-mags \
    --i-mags mags-derep.qza \
    --p-threads 4 \
    --p-seed 100 \
    --o-index mags-derep-index.qza \
    --verbose

Read mapping¶

The next step is read mapping—this time we will map the original reads to the dereplicated MAGs using the index generated in the previous step:

mosh assembly map-reads \
    --i-index mags-derep-index.qza \
    --i-reads reads.qza \
    --p-threads 2 \
    --p-seed 100 \
    --o-alignment-maps reads-to-mags-aln.qza \
    --parallel-config parallel.config.toml \
    --verbose

mosh assembly map-reads \
    --i-index mags-derep-index.qza \
    --i-reads reads.qza \
    --p-threads 2 \
    --p-seed 100 \
    --o-alignment-maps reads-to-mags-aln.qza \
    --verbose

MAG length estimation¶

To estimate MAG abundance, we need to find the length of each recovered genome so that we can normalize counts of reads based on genome length. To achieve that, you can use the get-feature-lengths action:

mosh annotate get-feature-lengths \
    --i-features mags-derep.qza \
    --o-lengths mags-derep-lengths.qza \
    --verbose

Abundance estimation¶

Finally, we are ready to estimate the abundance of our MAGs. The action we will use below supports two different kinds of metric which can be used as a proxy of genome abundance: RPKM (Reads Per Kilobase per Million mapped reads) and TPM (Transcripts Per Million)—for now we will continue with TPM but feel free to check out this post to learn more about those. We will also set minimal mapping quality to 42 to ensure we are taking into account only the reads which mapped to our MAGs perfectly.

mosh annotate estimate-abundance \
    --i-alignment-maps reads-to-mags-aln.qza \
    --i-feature-lengths mags-derep-lengths.qza \
    --p-metric tpm \
    --p-min-mapq 42 \
    --p-threads 4 \
    --o-abundances mags-abundances.qza \
    --verbose

Taxonomic composition visualization¶

Great! Finally, we can now combine our new feature table with the taxonomy obtained previously to visualize the abundances of the dereplicated MAGs:

mosh taxa barplot \
    --i-table mags-abundances.qza \
    --i-taxonomy mags-taxonomy.qza \
    --o-visualization mags-taxa-barplot.qzv \
    --verbose

Your visualization should look similar to this one.