Gene count vs Relative gene abundance

Question

[dpc]

Hi there!!! While reading to an article (Chatelier et. al., 2013) related to metagenomics, I have come accross two terminologies: Gene count and Gene richness. Are they same? They have stated

"we selected the threshold of one read for gene identification to include the rare genes into the analysis."

What does this threshold level refers to?

Can anyone please help me?

[Friederike]

can you provide the full reference or, even a better, a link to the publication you're mentioning?

My best bet is that gene richness is a proxy measure for the diversity of a person's microbiome.

[dpc]

Yes...they exactly done that. But, what if we do it from the relative abundance file of different taxa generated by MetaPhlAn? Will it be better way to do? This is link to the article

[Friederike]

The article you've linked is not by Chatelier et al and is not from 2013. You've also changed the question. What exactly do you want to know? What should MetaPhlAn be a better way for?

[dpc]

extremely sorry for the confusion. Chatelier et al ("gene counting" section under Methos) Look, firstly I have the confusion regarding the statement from the article:

we selected the threshold of one read for gene identification to include the rare genes into the analysis."

In addition, I have one more query. We can use the gene relative abundance to comapre the alpha diversity between the samples. Right? From, MetaPhlAn (running the metagenome reads directly) we don't get any information regarding gene. Rather, we get relative abundance of different taxa. So, if we directly get this information, what is the need to get gene count and then calculate alpha diversity?

[Friederike]

I am no metagenomics expert, but I could see how Metaphlan may simply have not been an established tool while Chatelier et al were working on their analyses. The first Metaphlan paper seems to have been published in 2012, not sure if people were aware of it beforehand. I'm sure there are other reasons, too, but you may want to search for the advantage of Metaphlan over gene abundance directly.

Answer 1

[Friederike]

What does this threshold level refers to?

I would interpret "threshold of one read for gene identification" so that a given gene had to have at least one sequenced read in order for it to be retained for downstream analyses. In other words, all genes for which at least one sequencing read was found were included in the gene counting step.

[dpc]

That means all the genes that were present in the sample was taken into account. Right?

[andres.firrincieli]

not really. They did not assembled the metagenome:

An average of 34.1 million paired-end reads were produced for each sample and, after removing human contamination (∼0.1%, on average), 19.9 ± 6.7 million reads were mapped at a unique position of the reference catalogue of 3.3 million genes, using SOAP2.21 (ref. 55) by allowing at most two mismatches in the first 35-bp region and 90% identity over the read sequence;