MammalHom is a tool for translating gene symbols in one mammalian species to their homologues in other mammalian species. It currently supports translations back and forth between human, mouse, and rat gene symbols. As described below, it is based on the NCBI HomoloGene and Gene databases. MammalHom was developed by John Newman, an M.D./Ph.D. student working in the laboratory of Alan Weiner in the Department of Biochemistry, University of Washington. It is based on code developed for the L2L Microarray Database that is used to convert mouse or rat lists of genes to human gene symbols for inclusion in the database.

MammalHom tries to find a suitable homologue for a given gene in three different ways.

First, it looks for a HomoloGene entry corresponding to the input gene symbol and species. If it finds one, it reports the corresponding symbol for the output species. This is the most reliable translation, and is denoted by "HomoloGene:SYMBOL" (where SYMBOL is the input gene symbol) if you chose to include the source of the match in your output (see below).

Second, it looks for a direct gene symbol match between the two species. The HomoloGene database is constantly expanding, and some well-known genes are not yet included. If MammalHom finds that the gene symbol from one species has an exact match in another species, it reports the match. This is usually appropriate, but rarely the same symbol may refer to non-homolgous genes in different species. Symbols are extracted from the NCBI Gene database. This method is denoted by "Symbol match:SYMBOL" in the output.

Finally, if both of the first two methods fail, MammalHom repeats them for known aliases of the input gene. Aliases are extracted from the NCBI Gene database. Ambiguous aliases, that may refer to more than one gene, are not used. This is a fairly reliable method of translating "common names" that are not official symbols but are in popular use anyway - such as p53 instead of TP53 or Waf1 instead of CDKN1A. Many common names are ambiguous, however. CSB, for example, usually refers to the ERCC6 gene (Cockayne syndrome Group B), but is also an alias for CSH2 (chorionic somatomammotropin B). If an alias was used to find a match, the alias is reported in the output as "HomoloGene:ALIAS" or "Symbol match:ALIAS".

Input to MammalHom, whether you upload a file or use the text box, should be a list of gene symbols separated by spaces, tabs or newlines. You can put any number of gene symbols on one line, as long as each is separated by one or more spaces or tabs. An uploaded file must be a plain-text file, not a Word or Excel document. You can create a plain-text file from within these programs by choosing "File:Save As", then "Format: Text (Tab-delimited)" in Excel or "Format: Text only" in Word.

1. "Sort output alphabetically". By default, MammalHom returns its results in the same order in which you entered your gene symbols. This can be useful if you have copied your list of symbols from a spreadsheet, for example, and wish to paste the results into the same spreadsheet. However, you can choose to have MammalHom sort the gene symbols alphabetically instead. It first reports the gene symbols for which matches were found, in alphabetical order; then reports the gene symbols for which no matches were found, in alphabetical order.

2. "Include source of match in output". If this box is checked, an addition column will be added to the output describing how MammalHom arrived at a match between the two gene symbols - by using the HomoloGene database, a direct match of symbols, or if it needed to look up an alias first. See How does MammalHom work? above for details.

3. "Include input gene symbols in output". Checking this box means MammalHom will report the gene symbol you submitted, in addition to its match (if one was found). Unchecking the box means only the matching symbol will be reported.