16S rRNA pipeline

I used this pipeline for 16s analysis during my master and thesis

Table of content

1. Data description
2. IlluminaUtils
   • Quality filtering (minoche et al.)
   • Merging
3. Swarm
4. Chimera removal
5. Taxonomic assignation
6. Complete table with OTUs and tax

Data description

About raw data Raw data are in the preprocess directory. They are already demultiplexed and unziped.

If the demultiplexed fastq were named according to the barcode and index. Samples were renamed using the barcode_index.csv and the following bash lines

cd preprocess

# Demultiplexed data are renamed according to their barcode/index
for line in `cat ../barcode_index.csv`
 do
 source=`echo $line | awk 'BEGIN{FS=";"}{print $1"_"$2}'`
 out=`echo $line | awk 'BEGIN{FS=";"}{print $3}'`
 cp `echo $source'_1_R1.fastq'` ./`echo $out'_R1.fastq'`
 cp `echo $source'_1_R2.fastq'` ./`echo $out'_R2.fastq'`
 done

IlluminaUtils

IlluminaUtils python scripts by Meren were used to do the quality checking and paired end merging

Config files must be created for quality checking and merging

cd preprocess

# Create the first required file: qual-config.txt
ls *.fastq| awk 'BEGIN{FS="_R"}{print $1}' | uniq | awk 'BEGIN{print "sample\tr1\tr2"}{print $0 "\t" $0 "_R1.fastq\t" $0"_R2.fastq"}' > qual-config.txt


# Create the second required file: merge-config.txt 
ls *.fastq| awk 'BEGIN{FS="_R"}{print $1}' | uniq | awk 'BEGIN{print "sample\tr1\tr2"}{print $0 "\t" $0 "-QUALITY_PASSED_R1.fastq\t" $0"-QUALITY_PASSED_R2.fastq"}' > merge-config.txt

Quality filtering (minoche et al.)

The quality filtering is made with IlluminaUtils using Minoche et al recommanded filtering parameters.

cd preprocess

# Generate .ini files with barcode (.....) and primer associated for both R1 and R2.
# Exact match will be kept, and barcode + primers will be trimmed during filtering.
iu-gen-configs qual-config.txt --r1-prefix ^.....CCAGCAGC[C,T]GCGGTAA. --r2-prefix CCGTC[A,T]ATT[C,T].TTT[A,G]A.T

# loop for quality filtering
for i in *.ini
 do
 iu-filter-quality-minoche $i
 done

Output name: *-QUALITY_PASSED_R1.fastq and *-QUALITY_PASSED_R2.fastq

Merging

The merging function is similar to the quality filtering

iu-gen-configs merge-config.txt --r1-prefix ^.....CCAGCAGC[C,T]GCGGTAA. --r2-prefix CCGTC[A,T]ATT[C,T].TTT[A,G]A.T

# loop for quality filtering
for i in *.ini
 do
 iu-merge-pairs $i
 done

Once the quality filtering and merging done, I move the new fasta files into the fasta_dir and rename to only the sample name

for i in `ls preprocess/ | grep MERGED | sed 's/\(^.*\)_MERGED/\1/'`; do cp preprocess/$i"_MERGED" fasta_dir/$i".fasta"; done

Swarm

OTUs clustering is performed by the swarm algorythm developped by Mahé. There is no need for a thereshold using this algorythm.

All sequences are now in a fasta file per sample but the letters are sometime uppercase or lowercase. To avoid any confusion for the upcoming dereplication and clustering, I prefer to change all sequences to lowercase:

for i in fasta_dir/*.fasta; do awk '{if(!/>/){print tolower($0)}else{print $0}}' $i > $i.temp | mv $i.temp $i ; done

Dereplication in each sample

Then we can dereplicate all sequences per sample. The idea is to have a fasta file with unique sequences and their frequences in the defline.

for i in `ls fasta_dir/ | grep .fasta | sed 's/\(^.*\).fasta/\1/'`; do vsearch --derep_fulllength fasta_dir/$i.fasta --sizeout --relabel_sha1 --fasta_width 0 --output swarm/dereplicate/$i.derep.fasta ; done

Vsearch counting format is >Seq;size=#; and it needs to be changed to >Seq_# for swarm

cd swarm/dereplicate/
sed -i 's/size=/_/' *.derep.fasta
sed -i 's/;//g' *.derep.fasta

At this point we can create a contingency table of the unique sequence per sample. The resulting table will have all unique sequences as rows and samples as columns and filled with the sequence abundance. Special thanks to Frederic Mahe for that code.

cd swarm/dereplicate/

awk 'BEGIN {FS = "[>_]"}
           # Parse the sample files
           /^>/ {contingency[$2][FILENAME] = $3
           amplicons[$2] += $3
           if (FNR == 1) {
               samples[++i] = FILENAME
           }
          }
    END {# Create table header
          printf "amplicon"
          s = length(samples)
          for (i = 1; i <= s; i++) {
              printf "\t%s", samples[i]
          }
          printf "\t%s\n", "total"

          # Sort amplicons by decreasing total abundance (use a coprocess)
          command = "LC_ALL=C sort -k1,1nr -k2,2d"
          for (amplicon in amplicons) {
               printf "%d\t%s\n", amplicons[amplicon], amplicon |& command
          }
          close(command, "to")
          FS = "\t"
          while ((command |& getline) > 0) {
              amplicons_sorted[++j] = $2
          }
          close(command)

          # Print the amplicon occurrences in the different samples
          n = length(amplicons_sorted)
          for (i = 1; i <= n; i++) {
               amplicon = amplicons_sorted[i]
               printf "%s", amplicon
               for (j = 1; j <= s; j++) {
                   printf "\t%d", contingency[amplicon][samples[j]]
               }
               printf "\t%d\n", amplicons[amplicon]
          }}' *derep.fasta > ../amplicon_contingency_table.csv

Dereplication for all sample

A dereplication at the sudy level is now required before the OTU clustering. It provides a unique fasta, with unique sequence and their abundance in the defline.

cd swarm/dereplicate/

export LC_ALL=C
cat *derep.fasta | \
awk 'BEGIN {RS = ">" ; FS = "[_\n]"}
     {if (NR != 1) {abundances[$1] += $2 ; sequences[$1] = $3}}
     END {for (amplicon in sequences) {
         print ">" amplicon "_" abundances[amplicon] "_" sequences[amplicon]}}' | \
sort --temporary-directory=$(pwd) -t "_" -k2,2nr -k1.2,1d | \
sed -e 's/\_/\n/2' > Temperature_effect.fasta

Swarm

Swarm was used with default value -d 1, meaning only the difference of 1 base pair is taken to compare sequences. Other parameters include -t 8 the number of processors used, -s -w -l and -o the various outputs and -f for ...

cd swarm/

swarm -d 1 -f -t 8 -s Temperature_effect.stat -w OTUs_rep.fasta -o Temperature_effect.swarm -l Temperature_effect.log dereplicate/Temperature_effect.fasta

Chimera removal

In this case, the chimera detection is done on the OTUs representative sequence only. According to Mahé, Swarm should create OTUs out of Chimera sequences rather than having them included in an OTU. A lot of processor time can be spared when the chimera detection is not done on the global fasta file.

The abundance format for each sequence needs to get from >Seq_# to >Seq;size=#; for Vsearch

cd swarm/

sed -i 's/_/;size=/' OTUs_rep.fasta
sed -i 's/^>.*/&;/' OTUs_rep.fasta

The chimera detection is done with Vsearch.

cd swarm/

vsearch --alignwidth 0 --uchime_denovo OTUs_rep.fasta --uchimeout Temperature_effect.uchimeout.txt