subsetting taxa

[apoosakkannu]

Hi, i wonder could it be possible to subset a object using specific taxa? Please let me know! thanks in advance!

[ChiLiubio]

Hi. Could you take a toy example?

[apoosakkannu]

Hi, Something like below example in phyloseq object, GP.chl = subset_taxa(GlobalPatterns, Phylum=="Chlamydiae")

[ChiLiubio]

Hi. Please try to use the following steps:

dataset$tax_table %<>% subset(Phylum=="Chlamydiae")
dataset$tidy_dataset()

I think the R subset function is enough to take the subset operation, so I do not add a function like subset_taxa of phyloseq package. I will consider your example. Thanks.

Chi

[apoosakkannu]

Thanks. It kind of worked, but i have problem when i do the differential abundance analysis. I have the following codes for my analysis in which when i subset the dataset to a specific taxa and converted into dataset1. But when do differential analysis, it is showing some other taxa too. I have attached dataset for your reference. Please give me a n idea where it goes wrong!

##upload all the data files

#upload the OTU abundance file abund_table<-read.csv("/scratch/project_2003832/pacbio_rawdata/bird_bat/stat/decontam_all_dataset/otu_table_decontam_all.csv",row.names=1,check.names=FALSE)

#upload the meta data file meta_table<-read.csv("/scratch/project_2003832/pacbio_rawdata/bird_bat/stat/decontam_all_dataset/sample_table_decontam_all.csv",row.names=1,check.names=FALSE)

#convert the meta data variables as factors col_names <- names(meta_table) meta_table[,col_names] <- lapply(meta_table[,col_names] , factor)

#upload the tree #require "ape" package library(ape) #version 5.5

OTU_tree <- read.tree("/scratch/project_2003832/pacbio_rawdata/bird_bat/stat/decontam_all_dataset/phylo_tree_decontam_all.tre")

#load the taxonomy OTU_taxonomy<-read.csv("/scratch/project_2003832/pacbio_rawdata/bird_bat/stat/decontam_all_dataset/tax_table_decontam_all.csv",row.names=1,check.names=FALSE)

make the taxonomic information unified, very important #require "tidytree" and "microeco" packages

library(tidytree) #version 0.3.5 library(microeco) #version 0.11.0

OTU_taxonomy %<>% tidy_taxonomy

##create and cleanup data object #The packages "microeco" and "tidytree" uploaded already

create a dataset of all the data uploaded

dataset <- microtable$new(sample_table = meta_table, otu_table = abund_table, tax_table = OTU_taxonomy, phylo_tree = OTU_tree)

#make the OTU and sample information consistent across all files in the dataset object dataset$tidy_dataset()

dataset

#calculate alpha diversity dataset$cal_alphadiv(PD = TRUE)

#calculate beta diversity dataset$cal_betadiv(unifrac = TRUE)

#calculate the abundance dataset$cal_abund()

save(dataset, file = "dataset.RData")

##select only Campylobacter taxa dataset1 <- clone(dataset) dataset1$tax_table %<>% subset(Genus == "g__Campylobacter") dataset1$tidy_dataset() dataset1

#identify the differential abundance taxa for 0.05% relative abundace t4 <- trans_diff$new(dataset = dataset1, method = "KW_dunn", group = "Host_taxa", taxa_level = "OTU", filter_thres = 0.005) t4$res_diff dataset.rdata.zip

[ChiLiubio]

Hi. You need perform dataset1$cal_abund() before the differential test. With this operation, the taxa_abund can be updated for the use of trans_diff.

[apoosakkannu]

Thanks, but it will be not original abundance right? because i subset only specific taxa! Please let me know if i can just plot taxa g__Campylobacter without any subset?

[ChiLiubio]

Hi. I got it! Please run the following steps and check whether it is correct.

# add OTU as a taxonomic level in tax_table
dataset$add_rownames2taxonomy(use_name = "OTU")
# cal_abund can return taxa_abund with the last data.frame named OTU
dataset$cal_abund()
# select some
dataset$taxa_abund$OTU %<>% .[grepl("g__Campylobacter", rownames(.)), ]
# run
 t4 <- trans_diff$new(dataset = dataset1, method = "KW_dunn", group = "Host_taxa", taxa_level = "OTU", filter_thres = 0.005)

[apoosakkannu]

Thanks. it worked.

[apoosakkannu]

Hi, I wonder could it be possible to subset some taxa without affecting their original abundance or count for the alpha and beta diversity analysis.

[ChiLiubio]

Hi. Could you please explain it with more details? I didnot get you.

[apoosakkannu]

I need to get the microeo obiect data only for following potential pathogenic genus, g__Acinetobacter|g__Aeromonas|g__Anaplasma|g__Bacillus|g__Bartonella|g__Borrelia|g__Campylobacter|g__Chlamydia|g__Citrobacter|g__Clostridiodes|g__Clostridium|g__Corynebacterium|g__Coxiella|g__Ehrilchia|g__Enterobacter|g__Enterococcus|g__Escherichia|g__Francisella|g__Klebsiella|g__Listeria|g__Mycobacterium|g__Mycoplasma|g__Neoehrilchia|g__Pasteurella|g__Proteus|g__Pseudomonas|g__Rickettsia|g__Salmonella|g__Serratia|g__Staphylococcus|g__Streptococcus|g__Vibrio|g__Yersinia'

and then proceed with normal analysis such as alpha and beta diversity.

On Wed, Nov 16, 2022 at 3:48 AM Chi Liu @.***> wrote:

Hi. Could you please explain it with more details? I didnot get you.

— Reply to this email directly, view it on GitHub https://github.com/ChiLiubio/microeco/issues/143#issuecomment-1316164465, or unsubscribe https://github.com/notifications/unsubscribe-auth/AMC3WZSIUBCZ66VBP7TE4HDWIQ4QBANCNFSM6AAAAAAQIWGD2U . You are receiving this because you authored the thread.Message ID: @.***>

[apoosakkannu]

actually something like here in the link, https://github.com/joey711/phyloseq/issues/1048

On Wed, Nov 16, 2022 at 9:05 AM Anbu Poosakkannu @.***> wrote:

I need to get the microeo obiect data only for following potential pathogenic genus, g__Acinetobacter|g__Aeromonas|g__Anaplasma|g__Bacillus|g__Bartonella|g__Borrelia|g__Campylobacter|g__Chlamydia|g__Citrobacter|g__Clostridiodes|g__Clostridium|g__Corynebacterium|g__Coxiella|g__Ehrilchia|g__Enterobacter|g__Enterococcus|g__Escherichia|g__Francisella|g__Klebsiella|g__Listeria|g__Mycobacterium|g__Mycoplasma|g__Neoehrilchia|g__Pasteurella|g__Proteus|g__Pseudomonas|g__Rickettsia|g__Salmonella|g__Serratia|g__Staphylococcus|g__Streptococcus|g__Vibrio|g__Yersinia'

and then proceed with normal analysis such as alpha and beta diversity.

On Wed, Nov 16, 2022 at 3:48 AM Chi Liu @.***> wrote:

Hi. Could you please explain it with more details? I didnot get you.

— Reply to this email directly, view it on GitHub https://github.com/ChiLiubio/microeco/issues/143#issuecomment-1316164465, or unsubscribe https://github.com/notifications/unsubscribe-auth/AMC3WZSIUBCZ66VBP7TE4HDWIQ4QBANCNFSM6AAAAAAQIWGD2U . You are receiving this because you authored the thread.Message ID: @.***>

[ChiLiubio]

Hi. You can directly manipulate the tax_table in your microtable object. Like this

library(microeco)
data(dataset)
newdata <- clone(dataset)
# use this if your_selected_genera is a vector with each genus as an element
newdata$tax_table <- subset(newdata$tax_table,  Genus %in% your_selected_genera)
# use this if all are in one string like you paste, use g__Acinetobacter|g__Aeromonas|g__Anaplasma as example
newdata$tax_table <- newdata$tax_table[grepl("g__Acinetobacter|g__Aeromonas|g__Anaplasma", newdata$tax_table$Genus), ]
# then trim
newdata$tidy_dataset()

[apoosakkannu]

Thanks. The subseeting of taxa worked. But it did not affect the alpha and beta diversity calculations. They remain the same as original data.

On Wed, Nov 16, 2022 at 1:32 PM Chi Liu @.***> wrote:

Hi. You can directly manipulate the tax_table in your microtable object. Like this

library(microeco) data(dataset) newdata <- clone(dataset)

use this if your_selected_genera is a vector with each genus as an element

newdata$tax_table <- subset(newdata$tax_table, Genus %in% your_selected_genera)

use this if all are in one string like you paste, use g__Acinetobacter|g__Aeromonas|g__Anaplasma as example

newdata$tax_table <- newdata$tax_table[grepl("g__Acinetobacter|g__Aeromonas|g__Anaplasma", newdata$tax_table$Genus), ]

then trim

newdata$tidy_dataset()

— Reply to this email directly, view it on GitHub https://github.com/ChiLiubio/microeco/issues/143#issuecomment-1316851803, or unsubscribe https://github.com/notifications/unsubscribe-auth/AMC3WZWT4UAC3JKRG6CKNKDWITA5TANCNFSM6AAAAAAQIWGD2U . You are receiving this because you authored the thread.Message ID: @.***>

[ChiLiubio]

You should rerun the diversity calculations for the new data.