#Extending colonic mucosal microbiome analysis - Assessment of colonic lavage as a proxy for endoscopic colonic biopsies

#Euan Watt, MBChB, BSc(Hons); Matthew R Gemmell, BSc(Hons), MSc; Susan H Berry, BSc; Mark Glaire, MBChB, BSc(Hons); Freda Farquharson, BSc, MSc; Petra Louis, BSc; PhD; Graeme Murray, MBChB, PhD; Emad M El-Omar, MBChB, BSc(Hons), MD; Georgina Louise Hold, PhD

#This document contains code used to QC and analyse data by mothur from the Raw 16S rRNA sequence data

#Produce contig file for mothur

ls * | sed 's/_L001_R1_001.fastq//g' | sed 's/_L001_R2_001.fastq//g'| uniq | while read x; do echo "${x} ${x}_L001_R1_001.fastq ${x}_L001_R2_001.fastq"; done > colonoscopy.makecontigsfile.txt

#Make contigs
make.contigs(file=colonoscopy.makecontigsfile.txt)

#Calculate summary information of contigs
summary.seqs(fasta=colonoscopy.makecontigsfile.trim.contigs.fasta)

#Screen sequences by: 
#Length with min and max of 400 and 500
#Max amount of ambigous bases as 0
#Max length of a homopolymer as 10
screen.seqs(fasta=colonoscopy.makecontigsfile.trim.contigs.fasta, group=colonoscopy.makecontigsfile.contigs.groups, minlength=400, maxlength=500, maxambig=0, maxhomop=10)

#Calculate summary information of screened contigs
summary.seqs(fasta=colonoscopy.makecontigsfile.trim.contigs.good.fasta)

#Unique sequences
unique.seqs(fasta=colonoscopy.makecontigsfile.trim.contigs.good.fasta)
Output File Names: 
colonoscopy.makecontigsfile.trim.contigs.good.names
colonoscopy.makecontigsfile.trim.contigs.good.unique.fasta

#Produce a count table to replace the names and groups file
count.seqs(name=colonoscopy.makecontigsfile.trim.contigs.good.names, group=colonoscopy.makecontigsfile.contigs.good.groups)

#Align sequences to SILVA database 
align.seqs(fasta=colonoscopy.makecontigsfile.trim.contigs.good.unique.fasta, reference=silva.nr_v119.align)

#Calculate summary information of aligned sequences
summary.seqs(fasta=colonoscopy.makecontigsfile.trim.contigs.good.unique.align, count=colonoscopy.makecontigsfile.trim.contigs.good.count_table)

#Screen sequences by:
#Start and end of sequences of 6288 and 25316
screen.seqs(fasta=colonoscopy.makecontigsfile.trim.contigs.good.unique.align, count=colonoscopy.makecontigsfile.trim.contigs.good.count_table, start=6388, end=25316, processors=4)

#Calculate summary information of screened aligned data
summary.seqs(fasta= colonoscopy.makecontigsfile.trim.contigs.good.unique.good.align, count= colonoscopy.makecontigsfile.trim.contigs.good.good.count_table)

#Filter sequences to remove gaps and overhangs
filter.seqs(fasta= colonoscopy.makecontigsfile.trim.contigs.good.unique.good.align, vertical=T,trump=.)

#Calculate summary info of filtered data
summary.seqs(fasta= colonoscopy.makecontigsfile.trim.contigs.good.unique.good.filter.fasta, count= colonoscopy.makecontigsfile.trim.contigs.good.good.count_table)

#Unique sequences
unique.seqs(fasta= colonoscopy.makecontigsfile.trim.contigs.good.unique.good.filter.fasta, count= colonoscopy.makecontigsfile.trim.contigs.good.good.count_table)

#Preculster sequences allowing for up to 4 differences between sequences
pre.cluster(fasta= colonoscopy.makecontigsfile.trim.contigs.good.unique.good.filter.unique.fasta, count= colonoscopy.makecontigsfile.trim.contigs.good.unique.good.filter.count_table, diffs=4)

#Chimera removal using uchime
chimera.uchime(fasta=colonoscopy.makecontigsfile.trim.contigs.good.unique.good.filter.unique.precluster.fasta, count=colonoscopy.makecontigsfile.trim.contigs.good.unique.good.filter.unique.precluster.count_table, dereplicate=f)           

#Remove sequences classified as chimeras by uchime
remove.seqs(accnos=colonoscopy.makecontigsfile.trim.contigs.good.unique.good.filter.unique.precluster.uchime.accnos, fasta=colonoscopy.makecontigsfile.trim.contigs.good.unique.good.filter.unique.precluster.fasta, count=colonoscopy.makecontigsfile.trim.contigs.good.unique.good.filter.unique.precluster.count_table)

#Calculate summary information of current data
summary.seqs(fasta= colonoscopy.makecontigsfile.trim.contigs.good.unique.good.filter.unique.precluster.pick.fasta, count= colonoscopy.makecontigsfile.trim.contigs.good.unique.good.filter.unique.precluster.pick.count_table)

#Classify sequences using the SILVA datbase
classify.seqs(fasta=current, count=current, reference=silva.nr_v119.align, taxonomy=silva.nr_v119.tax, cutoff=80, processors=4)

#Remove sequences classified as Chloroplast, Mitochondria, Archaea, Eukaryota or unknown
remove.lineage(fasta= colonoscopy.makecontigsfile.trim.contigs.good.unique.good.filter.unique.precluster.pick.fasta, count= colonoscopy.makecontigsfile.trim.contigs.good.unique.good.filter.unique.precluster.pick.count_table, taxonomy=colonoscopy.makecontigsfile.trim.contigs.good.unique.good.filter.unique.precluster.pick.nr_v119.wang.taxonomy, taxon=Chloroplast-Mitochondria-Archaea-Eukaryota-unknown)

#Produce Phylotype files
phylotype(taxonomy=colonoscopy.makecontigsfile.trim.contigs.good.unique.good.filter.unique.precluster.pick.nr_v119.wang.pick.taxonomy, count=colonoscopy.makecontigsfile.trim.contigs.good.unique.good.filter.unique.precluster.pick.pick.count_table)

make.shared(list=colonoscopy.makecontigsfile.trim.contigs.good.unique.good.filter.unique.precluster.pick.nr_v119.wang.pick.tx.unique_list.list, count=colonoscopy.makecontigsfile.trim.contigs.good.unique.good.filter.unique.precluste
r.pick.pick.count_table)

classify.otu(list=colonoscopy.makecontigsfile.trim.contigs.good.unique.good.filter.unique.precluster.pick.nr_v119.wang.pick.tx.unique_list.list,count=colonoscopy.makecontigsfile.trim.contigs.good.unique.good.filter.unique.precluster.pick.pick.count_table, taxonomy=colonoscopy.makecontigsfile.trim.contigs.good.unique.good.filter.unique.precluster.pick.nr_v119.wang.pick.taxonomy)

#Cluster sequences based on dissimilarity
cluster.split(fasta=colonoscopy.makecontigsfile.trim.contigs.good.unique.good.filter.unique.precluster.pick.pick.fasta, count=colonoscopy.makecontigsfile.trim.contigs.good.unique.good.filter.unique.precluster.pick.pick.count_table, taxonomy=colonoscopy.makecontigsfile.trim.contigs.good.unique.good.filter.unique.precluster.pick.nr_v119.wang.pick.taxonomy, splitmethod=classify, cutoff=4, processors=2)

#Produce a shared file for 0.03 OTUs
make.shared(list=current, count=colonoscopy.makecontigsfile.trim.contigs.good.unique.good.f
ilter.unique.precluster.pick.pick.count_table, label=0.03)                                          
                                                                                                    
#Classify OTUs
classify.otu(list= colonoscopy.makecontigsfile.trim.contigs.good.unique.good.filter.unique.precluster.pick.pick.an.unique_list.list, count= colonoscopy.makecontigsfile.trim.contigs.good.unique.good.filter.unique.precluster.pick.pick.count_table, taxonomy=colonoscopy.makecontigsfile.trim.contigs.good.unique.good.filter.unique.precluster.pick.nr_v119.wang.pick.taxonomy, label=0.03)

#Remove OTUs where there is less that 2 individuals across all the samples
remove.rare(list=colonoscopy.makecontigsfile.trim.contigs.good.unique.good.filter.unique.precluster.pick.pick.an.unique_list.list, count=colonoscopy.makecontigsfile.trim.contigs.good.unique.good.filter.unique.precluster.pick.pick.count_table, nseqs=2, bygroup=f, label=0.03)

#Make shared file for rare removed 0.03 OTUs
make.shared(list=colonoscopy.makecontigsfile.trim.contigs.good.unique.good.filter.unique.precluster.pick.pick.an.unique_list.0.03.pick.list, count= colonoscopy.makecontigsfile.trim.contigs.good.unique.good.filter.unique.precluster.pick.pick.pick.count_table, label=0.03)

#Classify OTUs
classify.otu(list=colonoscopy.makecontigsfile.trim.contigs.good.unique.good.filter.unique.precluster.pick.pick.an.unique_list.0.03.pick.list, count= colonoscopy.makecontigsfile.trim.contigs.good.unique.good.filter.unique.precluster.pick.pick.pick.count_table, taxonomy=colonoscopy.makecontigsfile.trim.contigs.good.unique.good.filter.unique.precluster.pick.nr_v119.wang.pick.taxonomy, label=0.03)

#Check amount of sequences in each sample
count.groups(shared=colonoscopy.makecontigsfile.trim.contigs.good.unique.good.filter.unique.precluster.pick.pick.an.unique_list.0.03.pick.shared)

#Producing phylotype file containing sequences after rare OTU removal
list.seqs(count=colonoscopy.makecontigsfile.trim.contigs.good.unique.good.filter.unique.precluster.pick.pick.pick.count_table)

get.seqs(accnos=colonoscopy.makecontigsfile.trim.contigs.good.unique.good.filter.unique.precluster.pick.pick.pick.accnos, taxonomy=colonoscopy.makecontigsfile.trim.contigs.good.unique.good.filter.unique.precluster.pick.nr_v119.wang.pick.taxonomy)

phylotype(count= colonoscopy.makecontigsfile.trim.contigs.good.unique.good.filter.unique.precluster.pick.pick.pick.count_table, taxonomy=colonoscopy.makecontigsfile.trim.contigs.good.unique.good.filter.unique.precluster.pick.nr_v119.wang.pick.pick.taxonomy)

make.shared(list=colonoscopy.makecontigsfile.trim.contigs.good.unique.good.filter.unique.precluster.pick.nr_v119.wang.pick.pick.tx.unique_list.list, count= colonoscopy.makecontigsfile.trim.contigs.good.unique.good.filter.unique.precluster.pick.pick.pick.count_table)

classify.otu(list=colonoscopy.makecontigsfile.trim.contigs.good.unique.good.filter.unique.precluster.pick.nr_v119.wang.pick.pick.tx.unique_list.list, count=colonoscopy.makecontigsfile.trim.contigs.good.unique.good.filter.unique.precluster.pick.pick.pick.count_table, taxonomy=colonoscopy.makecontigsfile.trim.contigs.good.unique.good.filter.unique.precluster.pick.nr_v119.wang.pick.pick.taxonomy)

##OTU data analysis

#Rarefaction analysis 
rarefaction.single(shared=colonoscopy.makecontigsfile.trim.contigs.good.unique.good.filter.unique.precluster.pick.pick.an.unique_list.0.03.pick.shared, calc=sobs, freq=100)

#Use subsample size of 11000 to keep all samples in analysis

#Alpha diversity analysis
summary.single(shared=colonoscopy.makecontigsfile.trim.contigs.good.unique.good.filter.unique.precluster.pick.pick.an.unique_list.0.03.pick.shared, calc=sobs-chao-shannon-invsimpson-coverage, subsample=11000)

#Calculate Jaccard, Bray-Curtis and Yue & Clayton dissimilarity distances
dist.shared(shared=current, calc=jclass-thetayc-braycurtis, subsample=11000)

#Produce trees based on Jaccard, Bray-Curtis and Yue & Clayton dissimilarity distances
tree.shared(phylip=colonoscopy.makecontigsfile.trim.contigs.good.unique.good.filter.unique.precluster.pick.pick.an.unique_list.0.03.pick.jclass.0.03.lt.ave.dist)

tree.shared(phylip=colonoscopy.makecontigsfile.trim.contigs.good.unique.good.filter.unique.precluster.pick.pick.an.unique_list.0.03.pick.braycurtis.0.03.lt.ave.dist)

tree.shared(phylip=colonoscopy.makecontigsfile.trim.contigs.good.unique.good.filter.unique.precluster.pick.pick.an.unique_list.0.03.pick.thetayc.0.03.lt.ave.dist)

#Principal coordinates analysis based on Jaccard, Bray-Curtis and Yue & Clayton dissimilarity distances
pcoa(phylip=colonoscopy.makecontigsfile.trim.contigs.good.unique.good.filter.unique.precluster.pick.pick.an.unique_list.0.03.pick.jclass.0.03.lt.ave.dist)

pcoa(phylip=colonoscopy.makecontigsfile.trim.contigs.good.unique.good.filter.unique.precluster.pick.pick.an.unique_list.0.03.pick.braycurtis.0.03.lt.ave.dist)

pcoa(phylip=colonoscopy.makecontigsfile.trim.contigs.good.unique.good.filter.unique.precluster.pick.pick.an.unique_list.0.03.pick.thetayc.0.03.lt.ave.dist)

#Non-metric multidimensional scaling based on Jaccard, Bray-Curtis and Yue & Clayton dissimilarity distances
nmds(phylip=colonoscopy.makecontigsfile.trim.contigs.good.unique.good.filter.unique.precluster.pick.pick.an.unique_list.0.03.pick.jclass.0.03.lt.ave.dist, mindim=3, maxdim=3)

nmds(phylip=colonoscopy.makecontigsfile.trim.contigs.good.unique.good.filter.unique.precluster.pick.pick.an.unique_list.0.03.pick.braycurtis.0.03.lt.ave.dist, mindim=3, maxdim=3)

nmds(phylip=colonoscopy.makecontigsfile.trim.contigs.good.unique.good.filter.unique.precluster.pick.pick.an.unique_list.0.03.pick.thetayc.0.03.lt.ave.dist, mindim=3, maxdim=3)

#Parsimony analysis comparing Biopsy and Lavage samples based on Jaccard, Bray-Curtis and Yue & Clayton dissimilarity distance tree
parsimony(tree=colonoscopy.makecontigsfile.trim.contigs.good.unique.good.filter.unique.precluster.pick.pick.an.unique_list.0.03.pick.jclass.0.03.lt.ave.tre, group=16sBxCL_FA_bx_design.txt, groups=all)

parsimony(tree=colonoscopy.makecontigsfile.trim.contigs.good.unique.good.filter.unique.precluster.pick.pick.an.unique_list.0.03.pick.braycurtis.0.03.lt.ave.tre, group=16sBxCL_FA_bx_design.txt, groups=all)

parsimony(tree=colonoscopy.makecontigsfile.trim.contigs.good.unique.good.filter.unique.precluster.pick.pick.an.unique_list.0.03.pick.thetayc.0.03.lt.ave.tre, group=16sBxCL_FA_bx_design.txt, groups=all)

#AMOVA analysis comparing Biopsy and Lavage samples based on Jaccard, Bray-Curtis and Yue & Clayton dissimilarity distances
amova(phylip=colonoscopy.makecontigsfile.trim.contigs.good.unique.good.filter.unique.precluster.pick.pick.an.unique_list.0.03.pick.jclass.0.03.lt.ave.dist, design=16sBxCL_FA_bx_design.txt)

amova(phylip=colonoscopy.makecontigsfile.trim.contigs.good.unique.good.filter.unique.precluster.pick.pick.an.unique_list.0.03.pick.braycurtis.0.03.lt.ave.dist, design=16sBxCL_FA_bx_design.txt)

amova(phylip=colonoscopy.makecontigsfile.trim.contigs.good.unique.good.filter.unique.precluster.pick.pick.an.unique_list.0.03.pick.thetayc.0.03.lt.ave.dist, design=16sBxCL_FA_bx_design.txt)

#Rename output
mv colonoscopy.makecontigsfile.trim.contigs.good.unique.good.filter.unique.precluster.pick.pick.an.unique_list.0.03.pick.jclass.0.03.lt.ave.tre.parsimony 16sBxCL_jclass_FA_bx.parsimony
mv colonoscopy.makecontigsfile.trim.contigs.good.unique.good.filter.unique.precluster.pick.pick.an.unique_list.0.03.pick.jclass.0.03.lt.ave.tre.psummary 16sBxCL_jclass_FA_bx.psummary
mv colonoscopy.makecontigsfile.trim.contigs.good.unique.good.filter.unique.precluster.pick.pick.an.unique_list.0.03.pick.braycurtis.0.03.lt.ave.tre.parsimony 16sBxCL_braycurtis_FA_bx.parsimony
mv colonoscopy.makecontigsfile.trim.contigs.good.unique.good.filter.unique.precluster.pick.pick.an.unique_list.0.03.pick.braycurtis.0.03.lt.ave.tre.psummary 16sBxCL_braycurtis_FA_bx.psummary
mv colonoscopy.makecontigsfile.trim.contigs.good.unique.good.filter.unique.precluster.pick.pick.an.unique_list.0.03.pick.jclass.0.03.lt.ave.amova 16sBxCL_jclass_FA_bx.amova
mv colonoscopy.makecontigsfile.trim.contigs.good.unique.good.filter.unique.precluster.pick.pick.an.unique_list.0.03.pick.braycurtis.0.03.lt.ave.amova 16sBxCL_braycurtis_FA_bx.amova
mv colonoscopy.makecontigsfile.trim.contigs.good.unique.good.filter.unique.precluster.pick.pick.an.unique_list.0.03.pick.thetayc.0.03.lt.ave.amova 16sBxCL_thetayc_FA_bx.amova
mv colonoscopy.makecontigsfile.trim.contigs.good.unique.good.filter.unique.precluster.pick.pick.an.unique_list.0.03.pick.thetayc.0.03.lt.ave.tre.parsimony 16sBxCL_thetayc_FA_bx.parsimony
mv colonoscopy.makecontigsfile.trim.contigs.good.unique.good.filter.unique.precluster.pick.pick.an.unique_list.0.03.pick.thetayc.0.03.lt.ave.tre.psummary 16sBxCL_thetayc_FA_bx.psummary

#Parsimony analysis comparing Participants based on Jaccard, Bray-Curtis and Yue & Clayton dissimilarity distance trees
parsimony(tree=colonoscopy.makecontigsfile.trim.contigs.good.unique.good.filter.unique.precluster.pick.pick.an.unique_list.0.03.pick.jclass.0.03.lt.ave.tre, group=16sBxCL_individual_design.txt, groups=all)

parsimony(tree=colonoscopy.makecontigsfile.trim.contigs.good.unique.good.filter.unique.precluster.pick.pick.an.unique_list.0.03.pick.braycurtis.0.03.lt.ave.tre, group=16sBxCL_individual_design.txt, groups=all)

parsimony(tree=colonoscopy.makecontigsfile.trim.contigs.good.unique.good.filter.unique.precluster.pick.pick.an.unique_list.0.03.pick.thetayc.0.03.lt.ave.tre, group=16sBxCL_individual_design.txt, groups=all)

#AMOVA analysis comparing Participants based on Jaccard, Bray-Curtis and Yue & Clayton dissimilarity distances

amova(phylip=colonoscopy.makecontigsfile.trim.contigs.good.unique.good.filter.unique.precluster.pick.pick.an.unique_list.0.03.pick.jclass.0.03.lt.ave.dist, design=16sBxCL_individual_design.txt)

amova(phylip=colonoscopy.makecontigsfile.trim.contigs.good.unique.good.filter.unique.precluster.pick.pick.an.unique_list.0.03.pick.braycurtis.0.03.lt.ave.dist, design=16sBxCL_individual_design.txt)

amova(phylip=colonoscopy.makecontigsfile.trim.contigs.good.unique.good.filter.unique.precluster.pick.pick.an.unique_list.0.03.pick.thetayc.0.03.lt.ave.dist, design=16sBxCL_individual_design.txt)

#Subsample OTU file to 11000 sequences per sample prior to LEfSe analysis
sub.sample(shared=colonoscopy.makecontigsfile.trim.contigs.good.unique.good.filter.unique.precluster.pick.pick.an.unique_list.0.03.pick.shared, size=11000)

#LEfSe analysis of subsampled OTU data
lefse(shared=colonoscopy.makecontigsfile.trim.contigs.good.unique.good.filter.unique.precluster.pick.pick.an.unique_list.0.03.pick.0.03.subsample.shared, design=16sBxCL_FA_bx_lefse_design.txt)

#Classify sequences with Greengenes database for PICRUSt analysis
classify.seqs(fasta=colonoscopy.makecontigsfile.trim.contigs.good.unique.good.filter.unique.precluster.pick.pick.fasta, count=colonoscopy.makecontigsfile.trim.contigs.good.unique.good.filter.unique.precluster.pick.pick.count_table, referencegg_13_8_99.fasta, taxonomy=gg_13_8_99.gg.tax, processors=4)

#Classify OTUs with Greengenes database
classify.otu(list=colonoscopy.makecontigsfile.trim.contigs.good.unique.good.filter.unique.precluster.pick.pick.an.unique_list.list, count=colonoscopy.makecontigsfile.trim.contigs.good.unique.good.filter.unique.precluster.pick.pick.count_table, taxonomy=colonoscopy.makecontigsfile.trim.contigs.good.unique.good.filter.unique.precluster.pick.pick.gg.wang.taxonomy, label=0.03)

#Create a shared file
make.shared(list=colonoscopy.makecontigsfile.trim.contigs.good.unique.good.filter.unique.precluster.pick.pick.an.unique_list.list, count=colonoscopy.makecontigsfile.trim.contigs.good.unique.good.filter.unique.precluster.pick.pick.count_table, label=0.03)

#Produce a biom file
make.biom(shared=colonoscopy.makecontigsfile.trim.contigs.good.unique.good.filter.unique.precluster.pick.pick.an.unique_list.shared, label=0.03, reftaxonomy=gg_13_8_99.gg.tax, constaxonomy=colonoscopy.makecontigsfile.trim.contigs.good.unique.good.filter.unique.precluster.pick.pick.an.unique_list.0.03.cons.taxonomy, picrust= GG_13_5_otuMapTable/97_otu_map.txt)

#PICRUSt analysis carried out in Huttenhower galaxy with biom file

#PICRUSt produced Biom file containing predicted metagenome converted into a readable format 
biom convert -i 16sBxCL.picrust_predicted_metagenome.biom  -o 16sBxCL.PICRUST_results.txt --to-tsv

#PICRUST predicted KEGG pathways converted into readable format
biom convert -i 16sBxCL.picrust_predicted_metagenome.KEGG_pathways.biom  -o 16sBxCL.KEGG_pathways.PICRUST_results.txt --to-tsv

#These readable format files were manually transformed into mothur compatible shared files to carry out analysis with mothur

#Count amount of sequences within created KEGG pathway shared file
count.groups(shared=16sBxCL.KEGG_pathways.shared.txt)

#Count amount of sequences within created predicted metagenome shared file
count.groups(shared=16sBxCL.PICRUST_predicted_metagenome.shared.txt)

#Calculate alpha diversity metrics of predicted metagenome
summary.single(shared=16sBxCL.PICRUST_predicted_metagenome.shared.txt, calc=sobs-chao-shannon-invsimpson-coverage, subsample=500000)

#Calculate Jaccard, Bray-Curtis and Yue & Clayton dissimilarity distances for predicted metagenome
dist.shared(shared=16sBxCL.PICRUST_predicted_metagenome.shared.txt, calc=jclass-thetayc-braycurtis, subsample=500000)

#Produce trees based on Jaccard, Bray-Curtis and Yue & Clayton dissimilarity distances for the predicted metagenome
tree.shared(phylip=16sBxCL.PICRUST_predicted_metagenome.shared.jclass.unique.lt.ave.dist)

tree.shared(phylip=16sBxCL.PICRUST_predicted_metagenome.shared.braycurtis.unique.lt.ave.dist)

tree.shared(phylip=16sBxCL.PICRUST_predicted_metagenome.shared.thetayc.unique.lt.ave.dist)

#Principal coordinates analysis based on Jaccard, Bray-Curtis and Yue & Clayton dissimilarity distances for the predicted metagenome
pcoa(phylip=16sBxCL.PICRUST_predicted_metagenome.shared.jclass.unique.lt.ave.dist) 

pcoa(phylip=16sBxCL.PICRUST_predicted_metagenome.shared.braycurtis.unique.lt.ave.dist)

pcoa(phylip=16sBxCL.PICRUST_predicted_metagenome.shared.thetayc.unique.lt.ave.dist)

#Non-metric multidimensional scaling based on Jaccard, Bray-Curtis and Yue & Clayton dissimilarity distances for the predicted metagenome
nmds(phylip=16sBxCL.PICRUST_predicted_metagenome.shared.jclass.unique.lt.ave.dist, mindim=3, maxdim=3)

nmds(phylip=16sBxCL.PICRUST_predicted_metagenome.shared.braycurtis.unique.lt.ave.dist, mindim=3, maxdim=3)

nmds(phylip=16sBxCL.PICRUST_predicted_metagenome.shared.thetayc.unique.lt.ave.dist, mindim=3, maxdim=3)

#Calculate alpha diversity metrics of predicted KEGG pathways
summary.single(shared=16sBxCL.KEGG_pathways.shared.txt, calc=sobs-chao-shannon-invsimpson-coverage, subsample=500000)

#Calculate Jaccard, Bray-Curtis and Yue & Clayton dissimilarity distances for KEGG pathways
dist.shared(shared=16sBxCL.KEGG_pathways.shared.txt, calc=jclass-braycurtis-thetayc, subsample=500000)

#Principal coordinates analysis based on Jaccard, Bray-Curtis and Yue & Clayton dissimilarity distances for KEGG pathways
pcoa(phylip=16sBxCL.KEGG_pathways.shared.jclass.unique.lt.ave.dist)

pcoa(phylip=16sBxCL.KEGG_pathways.shared.braycurtis.unique.lt.ave.dist)

pcoa(phylip=16sBxCL.KEGG_pathways.shared.thetayc.unique.lt.ave.dist)

#Non-metric multidimensional scaling based on Jaccard, Bray-Curtis and Yue & Clayton dissimilarity distances for KEGG pathways
nmds(phylip=16sBxCL.KEGG_pathways.shared.jclass.unique.lt.ave.dist, mindim=3, maxdim=3)

nmds(phylip=16sBxCL.KEGG_pathways.shared.braycurtis.unique.lt.ave.dist, mindim=3, maxdim=3) 

nmds(phylip=16sBxCL.KEGG_pathways.shared.thetayc.unique.lt.ave.dist, mindim=3, maxdim=3)