Commit 38a3e09f authored by Bastien Macé's avatar Bastien Macé
Browse files

lulu processing

parent 336f9827
OTUtable <- read.fasta(Aquarium_2.clustered.fasta)
matchlist <- read.table(match_list_Aquarium_2.txt)
## prepare the files needed for LULU processing
curated_results <- lulu(OTUtable, matchlist)
## LULU processing with the lulu R function
## shows the OTU names and their abundance after the curation
vsearch --usearch_global Aquarium_2.fasta --db Aquarium_2.fasta --self --id .84 --iddef 1 --userout match_list_Aquarium_2.txt -userfields query+target+id --maxaccepts 0 --query_cov .9 --maxhits 10
\ No newline at end of file
......@@ -302,4 +302,8 @@ For more details on this SWARM processing step, see the original publication [he
<a name="step24"></a>
### IV - 4 - SWARM + LULU processing step (Pipelines D)
For pipelines D, the same SWARM algorithm than in pipelines C was used, with an additional post-clustering step run thanks to the LULU algorithm.
LULU eliminates OTUs by flagging the alleged erroneous OTUs of more abundant OTUs and merges them. The algorithm requires an OTU match list to provide the pairwise similarity scores of the OTUs, with a minimum threshold of sequence similarity set at 84% as recommended by the authors. Only OTU pairs with a sequence similarity above 84% can then be interpreted as “parent” for the most abundant one and “daughter” for the other. Both OTU will possibly be merged provided that the co-occurrence pattern of the OTU pair among samples is higher than 95% and the abundance ratio between the “potential parent” and “potential daughter” is higher than a minimum ratio set by default as the minimum observed ratio.
For more details on LULU, see the original publication [here](
\ No newline at end of file
Supports Markdown
0% or .
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment