Advice needed about incorporating read depth

[RachBioHaz]

Hi, I have a dataset that has some inherent structural differences in the regions sampled that affect expected read depth. There is a correlation between the read depth with the region sampled. I am looking at the data in two ways: on a region-by-region basis and as a landscape as a whole.
My question is: What is the best way of adding read depth to the ANCOM-BC model for both approaches? I am fairly new to community analysis and R, so would appreciate some advice on the approach as well as the syntax for this package.

Cheers

[FrederickHuangLin]

Hi @RachBioHaz,

Generally, you do not have to worry about sequencing depth using ANCOM-BC since its methodology was developed to correct this bias.

If the sequencing depths are vastly different, you can put it in the formula argument, for instance, formula = "depth + group"; however, based on our simulations and real data applications, the results remained similar most of the time.

Best,
Huang

[RachBioHaz]

Thank you. I hoped I had understood the documentation correctly.CheersRacheleSent from my Galaxy -------- Original message --------From: Frederick Huang Lin ***@***.***> Date: 30/3/23 07:12 (GMT+10:00) To: FrederickHuangLin/ANCOMBC ***@***.***> Cc: Rachele Gore ***@***.***>, Mention ***@***.***> Subject: Re: [FrederickHuangLin/ANCOMBC] Advice needed about incorporating read depth (Discussion #159) Hi @RachBioHaz, Generally, you do not have to worry about sequencing depth using ANCOM-BC since its methodology was developed to correct this bias. If the sequencing depths are vastly different, you can put it in the formula argument, for instance, formula = depth + group; however, based on our simulations and real data applications, the results remained similar most of the time. Best, Huang —Reply to this email directly, view it on GitHub, or unsubscribe.You are receiving this because you were mentioned.Message ID: ***@***.***>