Overview

Genemaps main purpose is provide an easy approach for mapping (gene) identifers from one identifier type to another, which is unfortunately often not a trivial task. This may be useful for translating identifiers into a format that you are more familiar with (such as gene symbols instead of Ensembl IDs) or even for translating between different species (e.g., from mouse IDs to human IDs) to compare datasets or annotations between species.

Basic mappers

Mappings in genemap are implemented by Mapper classes, which interface with external data sources to perform different types of mappings. Currently, the following main mappers are supported:

  • EnsemblMapper - Translates identifiers using Ensembl via Biomart. Supports mapping between various types of gene identifiers that are present in Ensembl (gene symbols, Ensembl/Entrez identifiers, e.g.) and between different species (using orthologous genes).
  • MgiMapper - Translates identifers using the homology table provided by MGI. Supports mapping between gene symbols and Entrez identifers and between several specifies (using orthologous genes).
  • CustomMapper - Translates identifiers using a custom mapping provided by the user. The nature of the performed mapping depends on the provided mapping table.

Each of these Mappers can be used by instantiating the corresponding Mapper class and calling the mapping methods on the Mapper object (map_ids, map_dataframe and fetch_mapping). Alternatively, each mapper can be accessed using the functional interface of genemap (see Usage for more details).

Compound mappers

Besides the basic Mapper classes, genemap provides two additional classes that enable the creation of compound mappers (combinations of basic mappers):

  • ChainedMapper - Chains a series of mappers into a single mapping. For example, if two mappers are provided that map from identifier types a –> b and b –> c respectfully, these mappers are combined into a single mapper that maps from a –> c.
  • CombinedMapper - Combines the results of multiple mappers into a single mapper. This can be used to combine the mappings of two mappers with incomplete mappings (e.g., missing genes) into a single mapping combining the results of both mappers.

The compound mappers are also available in the functional interface, but can currently not be used directly from the command line.