search

ModelSEED / ModelSEEDDatabase

This repository contains the definitive copy of the biochemistry and metadata used to construct models using the ModelSEED/ProbAnno approach
Other
53 stars 38 forks source link

Tweaks to template deployability #22

Closed ctseto closed 8 years ago

ctseto commented 8 years ago

Reactions file parsed for organelle/compartment names: chloroplast, plastid, etc, and used to augment exclusion.

Human Reactions.tsv adjusted to include additional compartments, including carboxysome, golgi, lysosome, mitochondria, nucleus, endoplasmic reticulum, peroxisome.

Mycobacteria compartments now cytosol, extracellular and periplasm.

Microbial templates also cytosol, extracellular, periplasm.

Microbial, Human, Mycobacterial templates have rxn07191. However cpx31109 appears to no longer exist. To get these templates to work, cpx31109 is excised.

Microbial, Human, Mycobacterial templates were test-built on local environment, and appear to build successfully without problems.

Issue: Reactions.tsv was built using GramNegative template as a starting point. However, in those templates only reactions 1->14399 have associated complexes. Reactions beyond 14399 do not presently have associated complexes, thus type is gapfilling instead of conditional.

chian commented 8 years ago

@cseto Is there a plan to approach the complexes issue?

@mmundy Does this get resolved by latest ProbModelsEED integration?

Sent from my iPhone

On Oct 26, 2015, at 1:01 PM, ctseto notifications@github.com wrote:

Reactions file parsed for organelle/compartment names: chloroplast, plastid, etc, and used to augment exclusion.

Human Reactions.tsv adjusted to include additional compartments, including carboxysome, golgi, lysosome, mitochondria, nucleus, endoplasmic reticulum, peroxisome.

Mycobacteria compartments now cytosol, extracellular and periplasm.

Microbial templates also cytosol, extracellular, periplasm.

Microbial, Human, Mycobacterial templates have rxn07191. However cpx31109 appears to no longer exist. To get these templates to work, cpx31109 is excised.

Microbial, Human, Mycobacterial templates were test-built on local environment, and appear to build successfully without problems.

Issue: Reactions.tsv was built using GramNegative template as a starting point. However, in those templates only reactions 1->14399 have associated complexes. Reactions beyond 14399 do not presently have associated complexes, thus type is gapfilling instead of conditional.

You can view, comment on, or merge this pull request online at:

https://github.com/ModelSEED/ModelSEEDDatabase/pull/22

Commit Summary

Tweaks to template deployability File Changes

A Templates/Human/Compartments.tsv (10) M Templates/Human/Reactions.tsv (15444) A Templates/Microbial/Compartments.tsv (3) M Templates/Microbial/Reactions.tsv (15067) M Templates/Mycobacteria/BiomassCompounds.tsv (40) A Templates/Mycobacteria/Compartments.tsv (4) M Templates/Mycobacteria/Reactions.tsv (15518) A Templates/Plant/Compartments.tsv (3) Patch Links:

https://github.com/ModelSEED/ModelSEEDDatabase/pull/22.patch https://github.com/ModelSEED/ModelSEEDDatabase/pull/22.diff — Reply to this email directly or view it on GitHub.

cseto commented 8 years ago

Looks like there was a typo: @ctseto is your intended recipient. I'm @cseto.

On Tue, Oct 27, 2015 at 9:38 AM, chian notifications@github.com wrote:

@cseto Is there a plan to approach the complexes issue?

@mmundy Does this get resolved by latest ProbModelsEED integration?

Sent from my iPhone

On Oct 26, 2015, at 1:01 PM, ctseto notifications@github.com wrote:

Reactions file parsed for organelle/compartment names: chloroplast, plastid, etc, and used to augment exclusion.

Human Reactions.tsv adjusted to include additional compartments, including carboxysome, golgi, lysosome, mitochondria, nucleus, endoplasmic reticulum, peroxisome.

Mycobacteria compartments now cytosol, extracellular and periplasm.

Microbial templates also cytosol, extracellular, periplasm.

Microbial, Human, Mycobacterial templates have rxn07191. However cpx31109 appears to no longer exist. To get these templates to work, cpx31109 is excised.

Microbial, Human, Mycobacterial templates were test-built on local environment, and appear to build successfully without problems.

Issue: Reactions.tsv was built using GramNegative template as a starting point. However, in those templates only reactions 1->14399 have associated complexes. Reactions beyond 14399 do not presently have associated complexes, thus type is gapfilling instead of conditional.

You can view, comment on, or merge this pull request online at:

https://github.com/ModelSEED/ModelSEEDDatabase/pull/22

Commit Summary

Tweaks to template deployability File Changes

A Templates/Human/Compartments.tsv (10) M Templates/Human/Reactions.tsv (15444) A Templates/Microbial/Compartments.tsv (3) M Templates/Microbial/Reactions.tsv (15067) M Templates/Mycobacteria/BiomassCompounds.tsv (40) A Templates/Mycobacteria/Compartments.tsv (4) M Templates/Mycobacteria/Reactions.tsv (15518) A Templates/Plant/Compartments.tsv (3) Patch Links:

https://github.com/ModelSEED/ModelSEEDDatabase/pull/22.patch https://github.com/ModelSEED/ModelSEEDDatabase/pull/22.diff — Reply to this email directly or view it on GitHub.

— Reply to this email directly or view it on GitHub https://github.com/ModelSEED/ModelSEEDDatabase/pull/22#issuecomment-151562384 .

chian commented 8 years ago

@ctseto

Sent from my iPhone

On Oct 27, 2015, at 11:41 AM, cseto notifications@github.com wrote:

Looks like there was a typo: @ctseto is your intended recipient. I'm @cseto.

On Tue, Oct 27, 2015 at 9:38 AM, chian notifications@github.com wrote:

@cseto Is there a plan to approach the complexes issue?

@mmundy Does this get resolved by latest ProbModelsEED integration?

Sent from my iPhone

On Oct 26, 2015, at 1:01 PM, ctseto notifications@github.com wrote:

Reactions file parsed for organelle/compartment names: chloroplast, plastid, etc, and used to augment exclusion.

Human Reactions.tsv adjusted to include additional compartments, including carboxysome, golgi, lysosome, mitochondria, nucleus, endoplasmic reticulum, peroxisome.

Mycobacteria compartments now cytosol, extracellular and periplasm.

Microbial templates also cytosol, extracellular, periplasm.

Microbial, Human, Mycobacterial templates have rxn07191. However cpx31109 appears to no longer exist. To get these templates to work, cpx31109 is excised.

Microbial, Human, Mycobacterial templates were test-built on local environment, and appear to build successfully without problems.

Issue: Reactions.tsv was built using GramNegative template as a starting point. However, in those templates only reactions 1->14399 have associated complexes. Reactions beyond 14399 do not presently have associated complexes, thus type is gapfilling instead of conditional.

You can view, comment on, or merge this pull request online at:

https://github.com/ModelSEED/ModelSEEDDatabase/pull/22

Commit Summary

Tweaks to template deployability File Changes

A Templates/Human/Compartments.tsv (10) M Templates/Human/Reactions.tsv (15444) A Templates/Microbial/Compartments.tsv (3) M Templates/Microbial/Reactions.tsv (15067) M Templates/Mycobacteria/BiomassCompounds.tsv (40) A Templates/Mycobacteria/Compartments.tsv (4) M Templates/Mycobacteria/Reactions.tsv (15518) A Templates/Plant/Compartments.tsv (3) Patch Links:

https://github.com/ModelSEED/ModelSEEDDatabase/pull/22.patch https://github.com/ModelSEED/ModelSEEDDatabase/pull/22.diff — Reply to this email directly or view it on GitHub.

— Reply to this email directly or view it on GitHub https://github.com/ModelSEED/ModelSEEDDatabase/pull/22#issuecomment-151562384 .

— Reply to this email directly or view it on GitHub.

mmundy42 commented 8 years ago

Yes, and with the curation that is being done to the model templates.

ctseto commented 8 years ago

@chian Haven't had much luck finding a reference that links all reactions with all complexes. Complexes and roles are better related.

./Mappings/ComplexRoles.kegg.tsv describes the relationship between complexes (older cpx) and roles (fr), for those in the KEGG database. This file is akin to the Roles.tsv in ./Templates.

./SOLRDump/ComplexRoles.tsv also has a file describing the older complexes (cpx) and roles (psfr).

./Templates/Complexes.tsv lists complexes (cpx...), names (mscpx., cpx., pscpx.,) and associated roles(ftr); and ./Templates/Roles.tsv has roles (ftr), a reaction name string and additional aliases (msfr, psfr). Cross-referencing reaction name strings from Roles.tsv with the associated reaction names in reaction.master.tsv did not generate a large number of matches. It turns out that the reaction names in Roles are less processed than those in those in reactions.master.tsv; thus applying some processing (changes in letter case, elimination of special characters and EC number) may increase the match count.

cseto commented 8 years ago

Please remove @cseto from this discussion, thanks.

On Tue, Oct 27, 2015 at 12:30 PM, ctseto notifications@github.com wrote:

@chian https://github.com/chian Haven't had much luck finding a reference that links all reactions with all complexes. Complexes and roles are better related.

./Mappings/ComplexRoles.kegg.tsv describes the relationship between complexes (older cpx) and roles (fr), for those in the KEGG database. This file is akin to the Roles.tsv in ./Templates.

./SOLRDump/ComplexRoles.tsv also has a file describing the older complexes (cpx) and roles (psfr).

./Templates/Complexes.tsv lists complexes (cpx...), names (mscpx., cpx., pscpx.,) and associated roles(ftr); and ./Templates/Roles.tsv has roles (ftr), a reaction name string and additional aliases (msfr, psfr). Cross-referencing reaction name strings from Roles.tsv with the associated reaction names in reaction.master.tsv did not generate a large number of matches. It turns out that the reaction names in Roles are less processed than those in those in reactions.master.tsv; thus applying some processing (changes in letter case, elimination of special characters and EC number) may increase the match count.

— Reply to this email directly or view it on GitHub https://github.com/ModelSEED/ModelSEEDDatabase/pull/22#issuecomment-151618389 .