FMAP

Functional Mapping and Analysis Pipeline for metagenomics and metatranscriptomics studies

Some example results are available at the homepage: https://qbrc.swmed.edu/FMAP/.

Features

• FMAP provides a more sensible reference protein sequence database based on UniRef.

• Identification of differentially-abundant genes KEGG Orthology

• Mapping differentially-abundant genes to pathways and modules (KEGG Pathway and KEGG Module)

• Mapping differentially-abundant genes to operons (ODB (v3))

Requirements

• Perl - scripting language

• R - statistical computing

• Statistics::R - Perl interface with the R statistical program

  • Use CPAN to install the module

  perl -MCPAN -e 'install Statistics::R'
  

  • or download the source and compile manually

  wget 'http://search.cpan.org/CPAN/authors/id/F/FA/FANGLY/Statistics-R-0.34.tar.gz'
  tar zxf Statistics-R-0.33.tar.gz
  cd Statistics-R-0.33
  perl Makefile.PL
  make
  make test
  make install
  

• Mapping program providing BLASTX search of sequencing reads: DIAMOND or USEARCH

• Linux commands: wget, cat, sort

• Bio::DB::Taxonomy - Access to a taxonomy database (which is required only if you want to build a custom database.)

• XML::LibXML - Perl Binding for libxml2 (which is required only if you want to download genome sequences.)

Command usages

• FMAP_database.pl
  • Process
  • Input
    1. UniRef sequence identity (50, 90, or 100)
    2. (optional) NCBI taxonomy IDs (integer)
  • Require Bio::DB::Taxonomy.
  • The following data files will be downloaded through FTP connection. If you have a problem in the FTP connection, please download the files through another method and copy them into "FMAP_data" directory before executing "FMAP_database.pl" command.
    ftp://ftp.ncbi.nlm.nih.gov/pub/taxonomy/taxdump.tar.gz
    ftp://ftp.uniprot.org/pub/databases/uniprot/current_release/knowledgebase/idmapping/idmapping.dat.gz
    ftp://ftp.uniprot.org/pub/databases/uniprot/uniref/uniref100/uniref100.fasta.gz
    or ftp://ftp.uniprot.org/pub/databases/uniprot/uniref/uniref90/uniref90.fasta.gz
    or ftp://ftp.uniprot.org/pub/databases/uniprot/uniref/uniref50/uniref50.fasta.gz
  • Require HTTP connection for KEGG API.

Usage:   perl FMAP_database.pl [options] 50|90|100 [NCBI_TaxID [...]]

Options: -h       display this help message
         -s       switch database
         -r       redownload data

• FMAP_prepare.pl

Usage:   perl FMAP_prepare.pl [options]

Options: -h       display this help message
         -r       redownload data
         -m FILE  executable file path of mapping program, "diamond" or "usearch" [diamond]
         -k       download prebuilt KEGG files

• FMAP_assembly.pl
  • Process
    • Read mapping: nucleotide sequence alignment using BWA
    • ORF mapping: protein sequence alignment using DIAMOND
  • Input
    1. Prefix of output files
    2. De novo assembled sequences in FASTA format
       • A FASTA file can be generated by metagenome assemblers such as SPAdes and MetaVelvet.
       • A FASTA file containing target genome sequences can be input instead.
    3. Whole metagenomic/metatranscriptomic shotgun sequencing reads in FASTQ or FASTA format
       • Multiple read files can be specified.
       • Paired-end read files must be specified comma-separated like "input.R1.fastq,input.R2.fastq".
       • The read files can be compressed by gzip.
  • Output
    1. Prefix.region.abundance.txt (abundances of ORF regions mapping to KEGG orthologies)
    2. Prefix.abundance.txt (abundances of KEGG orthologies)

Usage:   perl FMAP_assembly.pl [options] output.prefix assembly.fasta [input.fastq|input.R1.fastq,input.R2.fastq [...]] > summary.txt

Options: -h       display this help message
         -A STR   prepared assembly prefix
         -B       input indexed sorted BAM file instead of FASTQ file
         -m FILE  executable file path of mapping program, "diamond" or "usearch" [diamond]
         -p INT   number of threads [1]
         -e FLOAT maximum e-value to report alignments [10]
         -t DIR   directory for temporary files [$TMPDIR or /tmp]
         -a FLOAT search acceleration for ublast [0.5]
         -C STR   codon and translation e.g. ATG=M [NCBI genetic code 11 (Bacterial, Archaeal and Plant Plastid)]
         -S STR   comma-separated start codons [GTG,ATG,CTG,TTG,ATA,ATC,ATT]
         -T STR   comma-separated termination codons [TAG,TAA,TGA]
         -l INT   minimum translation length [10]
         -c FLOAT minimum coverage [0.8]
         -q INT   minimum mapping quality [0]
         -s STR   strand specificity, "f" or "r"
         -P STR   contig prefix used for abundance estimation

• FMAP_assembly_centrifuge.pl
  • Require Centrifuge.
  • Input
    1. FMAP_assembly.region.txt (ORF regions mapping to KEGG orthologies generated by FMAP_assembly)
    2. De novo assembled sequences in FASTA format
    3. Centrifuge index filename prefix (minus trailing .X.cf)
  • Output: FMAP_assembly.region.taxon.txt (FMAP_assembly.region.txt including a column of NCBI taxonomy IDs (integer))

Usage:   perl FMAP_assembly_taxon.pl [options] FMAP_assembly.region.txt assembly.fasta centrifuge.index

Options: -h       display this help message
         -p INT   number of threads [1]

• FMAP_assembly_heatmap.pl
  • Require Bio::DB::Taxonomy.
  • Input: FMAP_assembly.abundance.txt (abundances generated by FMAP_assembly)
  • Output: HTML format of abundance heatmap table

Usage:   perl FMAP_assembly_heatmap.pl [options] [name=]FMAP_assembly.abundance.txt [...] > FMAP_assembly_heatmap.html

Options: -h       display this help message
         -c FILE  comparison output file including orthology and filter columns
         -f INT   HTML font size
         -w INT   HTML table cell width

• FMAP_assembly_operon.pl
  • Input: FMAP_assembly.region.txt (ORF regions mapping to KEGG orthologies generated by FMAP_assembly)
  • Output: FMAP_assembly_operon.txt (ODB (v3) known operons consisting of orthologies located together on an assembled contig/scaffold/transcript)

Usage:   perl FMAP_assembly_operon.pl [options] FMAP_assembly.region.txt > FMAP_assembly_operon.txt

Options: -h       display this help message
         -a       print single-gene operons as well

• FMAP_download_genome.pl
  • Input: NCBI taxonomy IDs (integer)
  • Output: FASTA file containing genome sequences
  • Require XML::LibXML.

Usage:   perl FMAP_download_genome.pl [options] NCBI_TaxID [...] > genome.fasta

Options: -h       display this help message
         -a       assembly instead of genome

• FMAP_download.pl

Usage:   perl FMAP_download.pl [options]

Options: -h       display this help message
         -m FILE  executable file path of mapping program, "diamond" or "usearch" [diamond]
         -k       download prebuilt KEGG files
         -x       download only KEGG files

• FMAP_mapping.pl
  • Input: whole metagenomic (or metatranscriptomic) shotgun sequencing reads in FASTQ or FASTA format
  • Output: best-match hits in NCBI BLAST ‑m8 (= NCBI BLAST+ ‑outfmt 6) format

Usage:   perl FMAP_mapping.pl [options] input1.fastq|input1.fasta [input2.fastq|input2.fasta [...]] > blastx_hits.txt

Options: -h       display this help message
         -m FILE  executable file path of mapping program, "diamond" or "usearch" [diamond]
         -p INT   number of threads [1]
         -e FLOAT maximum e-value to report alignments [10]
         -t DIR   directory for temporary files [$TMPDIR or /tmp]
         -a FLOAT search acceleration for ublast [0.5]

• FMAP_quantification.pl
  • Input: output of "FMAP_mapping.pl"
  • Output: abundances (RPKM) of KEGG orthologies
  • Output columns: KEGG Orthology ID, orthology definition, abundance (RPKM)

Usage:   perl FMAP_quantification.pl [options] blast_hits1.txt [blast_hits2.txt [...]] > abundance.txt

Options: -h       display this help message
         -c       use CPM values instead of RPKM values
         -i FLOAT minimum percent identity [80]
         -l FILE  tab-delimited text file with the first column having protein names and the second column having the sequence lengths
         -o FILE  tab-delimited text file with the first column having protein names and the second column having the orthology names
         -d FILE  tab-delimited text file with the first column having orthology names and the second column having the definitions
         -w FILE  tab-delimited text file with the first column having read names and the second column having the weights

• FMAP_table.pl
  • Input: outputs of "FMAP_quantification.pl"
  • Output: abundance table
  • Output columns: KEGG Orthology ID, orthology definition, abundance of sample1, abundance of sample2, ...

Usage:   perl FMAP_table.pl [options] [name1=]abundance1.txt [[name2=]abundance2.txt [...]] > abundance_table.txt

Options: -h       display this help message
         -c       use raw read counts (readCount|count) instead of RPKM values
         -d       use normalized mean depths (meanDepth/genome) instead of RPKM values
         -f       use fractions
         -n       do not print definitions
         -r       print ORF regions

• FMAP_comparison.pl
  • Input: output of "FMAP_table.pl", sample group information
  • Output: comparison test statistics for orthologies
  • Output columns: KEGG Orthology ID, orthology definition, log2 fold change, p-value, FDR-adjusted p-value, filter (pass or fail)

Usage:   perl FMAP_comparison.pl [options] abundance_table.txt control1[,control2[...]] case1[,case2[...]] [...] > orthology_test_stat.txt

Options: -h       display this help message
         -t STR   statistical test for comparing sample groups, "kruskal", "anova", "poisson", "quasipoisson", "metagenomeSeq" [kruskal]
         -f FLOAT fold change cutoff [2]
         -p FLOAT p-value cutoff [0.05]
         -a FLOAT FDR-adjusted p-value cutoff [1]

• FMAP_pathway.pl
  • Input: output of "FMAP_comparison.pl"
  • Output: pathways enriched in filter-passed orthologies
  • Output columns: KEGG Pathway ID, pathway definition, orthology count, coverage, p-value, KEGG Orthology IDs with colors
  • KEGG Orthology IDs with colors: input of KEGG Pathway mapping (http://www.kegg.jp/kegg/tool/map_pathway2.html)

Usage:   perl FMAP_pathway.pl [options] orthology_test_stat.txt > pathway.txt

Options: -h       display this help message

• FMAP_module.pl
  • Input: output of "FMAP_comparison.pl"
  • Output: modules enriched in filter-passed orthologies
  • Output columns: KEGG Module ID, module definition, orthology count, coverage, p-value, KEGG Orthology IDs with colors
  • KEGG Orthology IDs with colors: input of KEGG Pathway mapping (http://www.kegg.jp/kegg/tool/map_pathway2.html)

Usage:   perl FMAP_module.pl [options] orthology_test_stat.txt > module.txt

Options: -h       display this help message

• FMAP_operon.pl
  • Input: output of "FMAP_comparison.pl"
  • Output: operons consisting of filter-passed orthologies
  • Output columns: ODB (v3) known operon IDs, operon definition, log2 fold change, KEGG Orthology IDs, KEGG Pathway IDs

Usage:   perl FMAP_operon.pl [options] orthology_test_stat.txt > operon.txt

Options: -h       display this help message
         -a       print single-gene operons as well

• FMAP_plot.pl
  • Input: output of "FMAP_pathway.pl", "FMAP_module.pl", or "FMAP_operon.pl"
  • Output: PNG format image file of p-value plot

Usage:   perl FMAP_plot.pl [options] pathway.txt|module.txt|operon.txt plot.pdf

Options: -h       display this help message
         -w INT   plot width [12]
         -h INT   plot height [8]
         -l FLOAT plot left margin [20]
         -p FLOAT p-value cutoff [0.05]
         -c FLOAT coverage cutoff [0 for pathway, 1 for module and operons]
         -d       do not print definition

• FMAP_all.pl
  • Input: configuration table file
  • Input columns: group (control, ...), sample name, input file of "FMAP_mapping.pl"
  • Output: script file including all FMAP commands, all FMAP outputs

Usage:   perl FMAP_all.pl [options] input.config [output_prefix]

Options: -h       display this help message
         -s       generate a script, but not execute it
         -m FILE  mapping: executable file path of mapping program, "diamond" or "usearch" [diamond]
         -t INT   mapping: number of threads [1]
         -c STR   comparison: statistical test for comparing sample groups, "kruskal", "anova", "poisson", "quasipoisson", "metagenomeSeq" [kruskal]
         -f FLOAT comparison: fold change cutoff [2]
         -p FLOAT comparison: p-value cutoff [0.05]
         -a FLOAT comparison: FDR-adjusted p-value cutoff [1]

Command orders

• Use the prebuilt database (UniRef90 and bacteria/archaea/fungi)

  1. FMAP_download.pl
  2. FMAP_mapping.pl
  3. FMAP_quantification.pl
  4. FMAP_table.pl
  5. FMAP_comparison.pl
  6. FMAP_pathway.pl
  7. FMAP_module.pl
  8. FMAP_operon.pl

• Use a custom database (you can define UniRef and taxonomy.)

  1. FMAP_database.pl
  2. FMAP_prepare.pl
  3. FMAP_mapping.pl
  4. FMAP_quantification.pl
  5. FMAP_table.pl
  6. FMAP_comparison.pl
  7. FMAP_pathway.pl
  8. FMAP_module.pl
  9. FMAP_operon.pl

Citation

Kim J, Kim MS, Koh AY, Xie Y, Zhan X. "FMAP: Functional Mapping and Analysis Pipeline for metagenomics and metatranscriptomics studies" BMC Bioinformatics. 2016 Oct 10;17(1):420. PMID: 27724866