Updated readme for new options in fastp filtering

readme.md

@@ -31,7 +31,7 @@ The scripts and associated software are executed in Bash in a Linux environment.
 
 Example data are availabe in the SourceForge repository (https://sourceforge.net/projects/facepai/). Within the example_data folder there are also a subset of data from Sigsgaard et al. (2019), where FACEPAI is compared to the method used in that paper.
 
-The script FACEPAI first pooles the forward and reversed reads respectively, then utilizes fastp to filter the reads from ligators and low quality reads (standard setting is keeping reads with phred >=Q15). The pooled and filtered reads are then merged using vsearch. Primers and reads below minimum length are removed using cutadapt, and the sequences are then filtered again using vsearch to remove sequences with N:s. Sequences are then dereplicated before mOTU:s are clustered using swarm, and checking for chimeras is carried out using vsearch. The resulting mOTU:s are thereafter BLAST-searched against the database file using blastn.
+The script FACEPAI first filters the forward and reversed reads for each replicate while at the same time merging them utilizing fastp with paired ended base correction option to filter out low quality reads (standard setting is keeping reads with phred >=Q15, overlap difference limit 40%, overlap length require 10%). The individual replicates (if more than one replicate is available) are then pooled. Primers and reads below minimum length are removed using cutadapt, and the sequences are then filtered again using vsearch to remove sequences with N:s. Sequences are then dereplicated before mOTU:s are clustered using swarm, and checking for chimeras is carried out using vsearch. The resulting mOTU:s are thereafter BLAST-searched against the database file using blastn.
 
 The most time consuming part of this process is downloading the reference database. However, once that is accomplished, the process of first preparing the database using CaPReSe and thereafter using FACEPAI for pooling, merging, filtering and identifying eDNA sequences requires a minimum of hands on involvement. Multiple samples can be processed using the same reference database. Be aware that the reference sequences are saved locally, and are not automatically updated. It is recommended to have a periodical routine to update the data sources.