Feature Request: Implement scoring functionality in `variant-linker` to calculate meta scores based on the VEP annotations obtained.

[berntpopp]

Proposed Solution

• Develop a new function or module to calculate meta scores.
• This function should take VEP annotation data as input and apply a predefined or configurable scoring algorithm.
• Ensure that this feature can be easily toggled on or off by the user.

Acceptance Criteria

• [x] Function to calculate meta scores is implemented.
• [x] The calculation logic is well-documented.
• [x] Users can enable or disable this feature as needed.
• [ ] Unit tests are written to validate the feature's functionality.

[berntpopp]

This step should be build from two config files:

1. to extract the variables from the annotion VEP step. For example:

   [
   {
   "transcript_consequences": [
     {
       "impact": "impact_variant",
       "cadd_phred": "cadd_phred_variant",
     },
   ],
   "colocated_variants": [
     {
       "frequencies": {
         "T": {
           "gnomade": "gnomade_variant"
         }
       },
     }
   ],
   }
   ]

2. one config file for the formula build in steps

   [
   {
   "meta_score": cadd_phred_variant * 2 + gnomade_variant * 10
   }
   ]

[berntpopp]

Description

Implement a new scoring functionality within the variant-linker tool to calculate meta scores based on VEP annotations. This feature will enhance the tool's capabilities by providing users with additional insights through custom scoring algorithms.

Proposed Solution

1. Configuration Files

   • Variable Assignment Configuration: This file will define how to extract and assign variables from the VEP annotation data.
   • Formula Configuration: This file will define the formula used to calculate the meta scores.

2. Schema.org Style Configuration

   • Utilize JSON-LD schema definitions to create a flexible and standard way of defining the configurations.

3. Transformation and Conditional Logic

   • The configuration should allow transformations (e.g., setting values to 0 if non-existent) and conditional logic for variable assignments (e.g., setting a variable based on the presence of specific terms).

4. Integration with Variant-Linker

   • The new scoring functionality should be integrated into the variant-linker process and be enabled or disabled via a command-line option.
   • The computed scores should be added back to each relevant element in the variant array.

Detailed Plan

Configuration Files

1. Variable Assignment Configuration

   • Define how to extract variables from VEP annotations.
   • Example:

     [
     {
      "transcript_consequences": [
        {
          "impact": "impact_variant",
          "cadd_phred": "cadd_phred_variant"
        }
      ],
      "colocated_variants": [
        {
          "frequencies": {
            "T": {
              "gnomad": "gnomad_variant"
            }
          }
        }
      ]
     }
     ]

2. Formula Configuration

   • Define the formula for calculating the meta scores.
   • Example:

     [
     {
      "meta_score": "cadd_phred_variant * 2 + gnomad_variant * 10"
     }
     ]

Configuration Schema

1. Variable Assignment Configuration Schema

   • Define the JSON-LD schema for variable assignment configuration.
   • Example:

     {
     "@context": "https://schema.org/",
     "@type": "Configuration",
     "variables": {
      "transcript_consequences": {
        "impact": "impact_variant",
        "cadd_phred": "cadd_phred_variant"
      },
      "colocated_variants": {
        "frequencies": {
          "T": {
            "gnomad": "gnomad_variant"
          }
        }
      }
     }
     }

2. Formula Configuration Schema

   • Define the JSON-LD schema for formula configuration.
   • Example:

     {
     "@context": "https://schema.org/",
     "@type": "Configuration",
     "formulas": [
      {
        "meta_score": "cadd_phred_variant * 2 + gnomad_variant * 10"
      }
     ]
     }

Integration with Variant-Linker

1. Command-Line Option

   • Add a new command-line option to enable or disable the scoring functionality.
   • Example:

     const argv = yargs
     .option('scoring_config', {
      alias: 'sc',
      description: 'Path to the scoring configuration file',
      type: 'string'
     })
     // existing options...

2. Scoring Function Implementation

   • Implement the scoring function that reads the configuration files, applies the formulas, and adds the computed scores to the variant data.

[berntpopp]

Implementation of testing will be handled in #5