Hi everyone, I have a question regarding some samples I taxonomically classified (through HUMANn2 pipeline) with metaphlan2
One of the samples, for example, reports a 100% relative abundance for Escherichia genus.
Re-running metaphlan2 with the -t rel_ab_w_read_stats option revealed that the “estimated total number of reads from known clades” was 1621.
Generally, having 1621 reads all belonging to the Escherichia genus, I would consider it a reliable information. However, I have some doubt about the fact that my sample mapped poorly since only 15% of the reads were mapped by HUMANn2
Can I consider the 100% relative abundance of Escherichia a reliable estimation?
Thank you very much for any help you would give me.
Hello again, I try to give more information on my question. Here, I attach a couple of outputs from metaphlan2 on my reads.
The first ("_wCounts") contains the relative abundances of the clades (100% Escherichia), together with the estimated number of reads per clade (>=1621), calculated setting -t rel_ab_w_read_stats.
The second ("_readsMap") contains all the reads mapping on the markers with a known taxonomy. The number of reads here is 537. This was obtained by setting the -t reads_map option
What is the difference in these two files? Moreover, since the number of reads assigned to Escherichia genus is 4 in the “_readsMap” file, should I consider that the relative abundance 100% Escherichia was based only on these 4 reads (which are, by the way, 2x2 paired reads)?
B1F0m_metaphlan_bugs_list_readsMap.tsv (93.0 KB)
B1F0m_metaphlan_bugs_list_wCounts.tsv (977 Bytes)