Amplicon Suite Aggregator

Description

Aggregates the results from AmpliconSuite

• Takes in zip files (completed results of individual or grouped Amplicon Suite runs)
• Aggregates results
  • Packages results into a new file
  • Outputs an aggregated .html and .csv file of results.
• Result file (in .tar.gz) is the aggregated results of all individual AmpliconSutie runs. It can be directly loaded onto AmpliconRepository.
• Can also take additional files along with the upload, provided the directory they are in contains a file named AUX_DIR.

Parameters available on GenePattern Server

• Available at: https://genepattern.ucsd.edu/gp/pages/index.jsf?lsid=urn:lsid:genepattern.org:module.analysis:00429:4.1
• Amplicon Architect Results (required)
  • Compressed (.tar.gz, .zip) files of the results from individual or grouped Amplicon Architect runs.
• project_name (required)
  • Prefix for output .tar.gz. Result will be named: output_prefix.tar.gz
• Amplicon Repository Email
  • If wanting to directly transfer results to AmpliconRepository.org, please enter your email.
• run amplicon classifier
  • Option for users to re-run Amplicon Classifier.
• reference genome
  • Reference genome used for Amplicon Architect results in the input.
• upload only
  • If 'Yes', then skip aggregation / classification and upload file to AmpliconRepository as is.
• name map
  • A two column file providing the current identifier for each sample (col 1) and a replacement name (col 2). Enables batch renaming of samples.

Installation

• Option 1: Git Clone
  • Step 1: git clone https://github.com/genepattern/AmpliconSuiteAggregator.git
  • Step 2: Install python package dependencies from list below.
  • If running into dependency issues, please use the docker methods.
• Option 2: Docker
  • Step 1: docker pull genepattern/amplicon-suite-aggregator

Dependencies

• List of python package dependencies used: intervaltree, matplotlib, numpy, pandas, Pillow, requests, scipy, urllib3

Options when running locally

Amplicon Suite Aggregator related options

• -flist / --filelist: Text file with files to use (one per line)

  • Create an input_list.txt file in this format where each line is a filepath to compressed aa_results:

  • Use this file as the input to the -flist flag.

• --files: List of files or directories to use. Can specify multiple paths of (.tar.gz, .zip) of Amplicon Architect results.
• -o / --output_name: Output Prefix. Will be used as project name for Amplicon Repository upload.
• --name_map: A two column file providing the current identifier for each sample (col 1) and a replacement name (col 2). Enables batch renaming of samples.
• -c / --run_classifier (Yes, No): If 'Yes', then run Amplicon Classifier on AA samples. If Amplicon Classifier results are already included in inputs, then they will be removed and samples will be re-classified.
• --ref (hg19, GRCh37, GRCh38, GRCh38_viral, mm10): Reference genome name used for alignment, one of hg19, GRCh37, GRCh38, GRCh38_viral, or mm10.

AmpliconRepository related options

• -u / --username: Email address for Amplicon Repository. If specified, will trigger an attempt to upload the aggregated file to AmpliconRepository.org.
• --upload_only (Yes, No): If 'Yes', then skip aggregation / classification and upload file to AmpliconRepository as is.
• -s / --server (dev, prod, local-debug): Which server to send results to. Accepts 'dev' or 'prod' or 'local-debug'. 'prod' is what most users want. 'dev' and 'local-debug' are for development and debugging.

How to run

• If using local CLI:

  • Running aggregator: python3 /path/to/AmpliconSuiteAggregator/path/src/AmpliconSuiteAggregator.py **options**
  • Using API to upload directly to AmpliconRepository without aggregation: python3 /path/to/AmpliconSuiteAggregator/path/src/AmpliconSuiteAggregator.py --files /path/to/aggregated/file.tar.gz -o projname -u your.amplicon.repository.username@gmail.com --upload_only Yes -s prod
    • Log into Amplicon Repository, you should see a new project with the output prefix you specified.

• If using Docker:

  • To Running aggregator:
    1. docker run --rm -it -v PATH/TO/INPUTS/FOLDER:/inputs/ genepattern/amplicon-suite-aggregator python3 /opt/genepatt/AmpliconSuiteAggregator.py **options**
  • Using API to upload directly to AmpliconRepository without aggregation:
    1. docker run --rm -it -v PATH/TO/INPUTS/FOLDER:/inputs/ genepattern/amplicon-suite-aggregator python3 /opt/genepatt/AmpliconSuiteAggregator.py -flist /path/to/input_list.txt -u YOUR_AMPLICON_REPOSITORY_EMAIL -o projname -u your.amplicon.repository.username@gmail.com --upload_only Yes -s prod
    2. Log into Amplicon Repository, you should see a new project with the output prefix you specified.

Programming Language

• Python

Contact

For any issues

• Edwin Huang, edh021@cloud.ucsd.edu