Gene name normalization (GNN)

This little tool helps to reduce a list of gene names: by normalizing each name, we combine identical genes to get a single entry. To do so, we remove certain parts from gene names: "p53 gene", "P53 variants", "p53 mutants", "human P53", and "P53 mRNA" will all get mapped to the gene "p53". In addition, we check for synonyms within the provided list (mostly abbreviations as in the example below). Note that we map certain family members to the general family name, if this also occurs as an entry in the list: "tumor necrosis factor alpha" will be combined with "tumor necrosis factor", and "IL-1 family members" with "IL-1".
For maximum effect, you can opt to include lexica (lists of gene names) in the search; this helps to map additional gene names to a single entry ("p53", "TP53", "tumor protein p53", "LFS1"). Sometimes it is useful to select all lexica, even if your genes do not originate from all of them. If you chose at least one lexicon, we can provide EntrezGene identifiers for each gene (for which we could assign one with reasonable certainty).
The normal output is a reduced list of gene names; if you want to see the original names for each entry and EntrezGene identifiers, please uncheck the box "Show normalized names only". To get a simple example for all this, keep the four initial entries ("EGFR" etc.), check the box for the human lexicon, and uncheck "Show normalized names only". You will see that all four names have been mapped to a single gene name, "EGFR", and that the EntrezGene ID for this gene (human version) is 1956.
Checking the box "Do not collapse list" will result in a list of your original entries (same size), where each name is mapped to a normalized version and a list of potential EntrezGene IDs. Note that this option overrides the "Show normalized names only" option.

A comparable tool is available as GNAT; this tool takes into account the context (that is, a piece of text) where a gene name occurs in, to map a gene to its EntrezGene identifier.