aeDNA example workflow

A possible workflow for your ancient environmental DNA sample could be:

wget https://github.com/grenaud/vgan/raw/main/test/input_files/euka/test.fq.gz

vgan euka -fq1 test.fq.gz --outFrag 

The taxon of bears (Ursidae) has been detected. We plot the taxon to verify that it looks ancient:

./visualize_detected_taxa.sh euka_output

After the verification, we extract the reads mapped to the bear taxon to do species detection:

less -S euka_output_FragNames.tsv | sed '3!d' | sed 's/\t/\n/g' | seqtk subseq test.fq.gz - | gzip > Ursidae.fq.gz

We extract the subgraph for the bears from our soibean graph:

./make_graph_files.sh Ursidae 

The script extracts the bear subgraph and creates all necessary files for mapping and analysing with soibean:

vgan soibean -fq1 Ursidae.fq.gz --dbprefix Ursidae -t 20 -o soibeanOut 

Once soibean is done, we can plot our results. First, we check the k-curve, which shows the number of sources best describing our data:

Rscript soibeanPlotk.R soibeanOut

Afterwards, we can choose the best run of our MCMC and plot the distribution on the phylogenetic tree (the tree files can be found: ~/vgan/share/vgan/soibean_dir/tree_dir.:

Rscript soibeanPlots.R Ursidae.new.dnd soibeanOutResult10.mcmc