/amr_pipeline

Primary LanguagePython

amr_pipeline

1. We start with raw fastq files (pair end)

  • From sequencing center
  • We need to do quality check. Criteria are length (if less than 75bp), remove bases of low quality, remove adapters. These can result in wrong interpretation downstream
  • Remove reads that map to the human genome
  • Output: quality controlled fastq

2. Run Metaphlan2 to get relative abundance of taxonomy

  • Input: quality controlled fastq
  • Estimating taxonomy abundance from metagenomes.
  • Cheng has a script to parse the Metaplhan outputs into different level and 'OTU' table
  • Output: 'OTU' table

3. Antibiotic resistance inference

  • Input: quality controlled fastq + reference database in fasta file (from SARG); metadata is a text file (ask Anni)
  • ShortBRED quantifies antibiotic resistance gene abundance based on hits to unique marker sequences
  • Metadata connects variants to genes to antibiotic class
  • Output here is a ShortBRED quantification file that gives you normalized abundance (reads per kilobase million) --> can be converted into a table