Skip to content

k8hertweck/FieldLabComparison

Repository files navigation

FieldLabComparison

Scripts for microbiome analysis of bacterial 16S sequences from ants.

Data:

  • SRA accessions:
    • PRJNA170251 (Mycocepurus): metadata in SRA correct, data/SraRunTableSRP018247.txt
    • PRJNA170250 (Trachymyrmex/Cyphomyrmex): metadata in SRA is not correct, revised metadata in data/SraRunTableSRP018246corrected.txt
  • Silva database:
    • Use Full length sequences and taxonomy references for version 128, which includes taxonomy file (these are downloaded as a part of the workflow below). Another option is to use the silva.bacteria.fasta which is a silva reference file from the MiSeq SOP. The silva.bacteria.fasta file can be trimmed to the V1-V3 region using the align.seqs() and the E.coliV1V3.fas file.

Dependencies:

Setting up data

  • dataDownload.sh download data from NCBI SRA and Silva database (takes several minutes depending on internet connectivity)
    • data/dataCheck.sh check SRA data against archived sequence files, do not need to run again

Mothur Workflow

  • MothuR analysis based on 454 and MiSeq tutorials to create OTU table
  • mothur_SilvaRef.bat create custom Silva alignment for reference (takes ~10 minutes)
  • mothur_prep.sh split, trim, and aggregate sequence files (takes several minutes)
  • mothur_otu.bat processes combined sequences and outputs OTU table

QIIME 1 Workflow

  • QIIME 1 analysis based on the 454 tutorial to create OTU table
  • qiime_workflow.sh uses 3 scripts to take fastq files and obtain an OTU table biom file
    • convert_fastaqual_fastq.py converts fastq to fasta & qual files in qiime
    • add_qiime_labels.py combines all fasta files into one fasta file in qiime
      • metadata_mapping_file.txt contains data that assist in combining all fasta files into one fasta file in qiime
    • pick_de_novo_otus.py a series of 7 scripts that outputs OTU table biom file
      • qiime_parameters.txt changes default settings of python script

Hypothesis testing and data visualization

  • R package phyloseq