gempro.py 文件源码

def kegg_mapping_and_metadata(self, kegg_organism_code, custom_gene_mapping=None, outdir=None,
                                  set_as_representative=False, force_rerun=False):
        """Map all genes in the model to KEGG IDs using the KEGG service.

        Steps:
            1. Download all metadata and sequence files in the sequences directory
            2. Creates a KEGGProp object in the protein.sequences attribute
            3. Returns a Pandas DataFrame of mapping results

        Args:
            kegg_organism_code (str): The three letter KEGG code of your organism
            custom_gene_mapping (dict): If your model genes differ from the gene IDs you want to map,
                custom_gene_mapping allows you to input a dictionary which maps model gene IDs to new ones.
                Dictionary keys must match model gene IDs.
            outdir (str): Path to output directory of downloaded files, must be set if GEM-PRO directories
                were not created initially
            set_as_representative (bool): If mapped KEGG IDs should be set as representative sequences
            force_rerun (bool): If you want to overwrite any existing mappings and files

        """

        # First map all of the organism's KEGG genes to UniProt
        kegg_to_uniprot = ssbio.databases.kegg.map_kegg_all_genes(organism_code=kegg_organism_code, target_db='uniprot')

        successfully_mapped_counter = 0

        for g in tqdm(self.genes):
            if custom_gene_mapping:
                kegg_g = custom_gene_mapping[g.id]
            else:
                kegg_g = g.id

            # Download both FASTA and KEGG metadata files
            kegg_prop = g.protein.load_kegg(kegg_id=kegg_g, kegg_organism_code=kegg_organism_code,
                                            download=True, outdir=outdir, set_as_representative=set_as_representative,
                                            force_rerun=force_rerun)

            # Update potentially old UniProt ID
            if kegg_g in kegg_to_uniprot.keys():
                kegg_prop.uniprot = kegg_to_uniprot[kegg_g]
                if g.protein.representative_sequence:
                    if g.protein.representative_sequence.kegg == kegg_prop.kegg:
                        g.protein.representative_sequence.uniprot = kegg_to_uniprot[kegg_g]

            # Keep track of missing mappings - missing is defined by no available sequence
            if kegg_prop.sequence_file:
                successfully_mapped_counter += 1

            log.debug('{}: loaded KEGG information for gene'.format(g.id))

        log.info('{}/{}: number of genes mapped to KEGG'.format(successfully_mapped_counter, len(self.genes)))
        log.info('Completed ID mapping --> KEGG. See the "df_kegg_metadata" attribute for a summary dataframe.')