Analysis of module extension

[GwennyGit]

• Investigation of overlaps between the gapfill module in refineGEMs and the extension module in SPECIMEN
• Analysis of the functions

---

The following points[1] were found:

• 🟣🟠 identify_missing_genes $\rightarrow$ Move to refineGEMs?/ Similar to compare_gene_list & get_missing_genes2reactions
• 🟣 find_best_hits $\rightarrow$ Similar to/Could be used in PGAB-Pipeline; Overlaps with/similar to read_tsv
• map_to_kegg_row
  $\rightarrow$ 🟠 More general question: Does the locus tag look-up work this way?
  $\rightarrow$ 🟣 Similar to look-up in get_kegg_genes, get_locus_ec & get_locus_ec_kegg $\rightarrow$ Merge all into one function? Different set-ups for the locus tag 🤔
  $\rightarrow$ 🟣 Mapping to MetaCyc/MetaNetX? $\rightarrow$ biocyc API?
• 🟠 add_gene $\rightarrow$ Redundant when compared to create_gp?
• 🟠 build_metabolite_mnx
  $\rightarrow$ Mapping to ID $\rightarrow$ Then addition?
  $\rightarrow$ What is the purpose of this function?
  $\rightarrow$ Similar to create_species?
• 🔴🟠 build_metabolite_kegg $\rightarrow$ Lines 762 - 766 redundant
• 🟠 get_metabolites_mnx $\rightarrow$ Does the calculation really work in line 918?
• 🟣 Merge/transfer overlaps into a helper function for all build_metabolites functions?
• 🟣 Merge/transfer overlaps into a helper function for all get_metabolites functions?
• 🟠 get_metabolites_kegg $\rightarrow$ Why add _i to metabolite id?
• 🟠 add_reaction $\rightarrow$ Similar to create_reaction?
• 🟠 Answer to question here $\rightarrow$ No, the compartment does not have to be in the metabolite name as long as each metabolite has its unique id. If create_random_id creates a unique id for atp_c and another for atp_e, there should be no issue.
• 🟠 In line 1086 $\rightarrow$ Do you want to add there the KEGG ID or the MetaNetX one?
• 🟣 Rename function in line 1200 to extend_model?
• 🟣🟠 Instead of get_reaction_annotation_dict, maybe compare all annotations?/ Search all reactions with same annotations?
• 🟣 run $\rightarrow$ Add printout before MEMOTE?

---

Overall ideas:

• 🟣🟠 Reorder addition of new entities? Order similar to how it is done in gapfill_model/ Change set-up?
• 🟣🟠 extent_model $\rightarrow$ Similar to idea in refineGEMs for filling gaps via GFF file
• 🟣🟠 Could be transferred to refineGEMs and adjusted for the gap-filling part

---

[1]
Legend: 🟠 open for discussion, 🟣 idea, 🔴 problem, 🟤 set to the side, ⚫ removed