Input and Output Files

This page details the type and format of the files VIOLIN takes as input, as well as the format of VIOLIN output

Input

As VIOLIN is developed as part of a larger modeling framework, the default files expected are model spreadsheets in the BioRECIPE format, and a machine reading output spreadsheet:

Model input spreadsheet

Element Name

Element ID

Element Type

Variable

Positive Regulators

Negative Regulators

Name

ID

Type

variable name

pos regulators list

neg regulators list

The minimum required information from a model is: element/target name, element/target type, element/target ID, positive and negative regulator/source names, types, and IDs. The nomenclature does not have to be specific, as VIOLIN has the capabilities to recognize all of the commonly used vocabulary.

LEE input spreadsheet

Element Name

Element ID

Element Type

Positive Reg Name

Positive Reg ID

Positive Reg Type

Negative Reg Name

Negative Reg ID

Negative Reg Type

PMCID

Evidence

Name

ID

Type

Pos regulator

reg ID

reg type

Neg regulator

reg ID

reg type

PMCID

evidence text

The minimum information required from reading output is: element/target name, element/target type, element/target ID, positive and negative regulator/source names, types, and IDs. The nomenclature does not have to be specific, as VIOLIN has the capabilities to recognize all of the commonly used vocabulary.

Additionally, VIOLIN has built-in functions to allow for other input file formats, found on the Formatting (VIOLIN.formatting) page.

Currently accepted file types are comma-separated files (.csv), tab-separated files (.txt or .tsv), and excel spreadsheets (.xlsx).

Output

VIOLIN output is sorted into five .csv output files:

  • Total output - all interactions listed by total score

  • Corroborations - all interactions classified as corroborations, listed by total score

  • Extensions - all interactions classified as extensions, listed by total score

  • Contradictions - all interactions classified as contradictions, listed by total score

  • Questionable - all interactions classified as questionable, listed by total score