manual-curation

Manual trait name curation protocol

Introduction

Detailed description of the current approach can be found in the addendum. This protocol, which is currently performed with TSV files and spreadsheets, is planned to be replaced by a web interface for manual curation.

Protocol overview

The protocol consists of four parts which are done in sequence by different people. Some parts are “technical” and are performed by a developer, other are “biological” and are performed by a curator.

1. Fetch data (technical). The latest ClinVar data is downloaded and the trait names are extracted. They are attempted to be automatically mapped to ontology terms using ZOOMA. The traits which cannot be mapped automatically are output as a separate file, which is loaded into a Google spreadsheet.
2. Curate (biological). The curator goes through the spreadsheet and fills it in, performing manual curation. Other people review the results and comment on them.
3. Extract results (technical). Curation results are extracted from the spreadsheet into a TSV file. Some accompanying data is prepared for providing feedback to EFO.

Setting up environment

To follow the technical steps of the protocol, you will need to set up the environment. You can follow the installation instructions in the main README or set up for Open Targets processing as below.

First, set up the common environment as explained in the build instructions.

Next, set up the protocol-specific environment, filling in ${CURATION_RELEASE}:

# Identifier of the current manual curation iteration, which is just the current date.
# Be sure to write this down and to set it to the same value in later parts of this protocol.
export CURATION_RELEASE=YYYY-MM-DD
export CURATION_RELEASE_ROOT=${BATCH_ROOT_BASE}/manual_curation/${CURATION_RELEASE}

Review checklist

• The mappings selected for each trait are adequate
• Good/bad criteria for curation are observed (see the manual curation protocol, section “Criteria to manually evaluate mapping quality”)
• The number of traits in the finished_mappings_curation.tsv file is the same as in the spreadsheet after applying all relevant filters
• Important: spreadsheet does not contain line endings, or extraneous space symbols, in trait names (can be checked by a regexp search)
• For submitting terms to EFO
  • Terms for import all have associated HP or MONDO IDs
  • Information for new terms has been populated for as many traits as possible
  • GitHub issue has been created and linked in the issue