search

ModelSEED / Model-SEED-core

The Model SEED is a tool for building, curating, and analyzing gonome-scale metabolic models. Visit the Model SEED homepage for installation instructions and full feature documentation.
http://bionet.mcs.anl.gov/index.php/Model_SEED_Homepage
Other
19 stars 10 forks source link

Autocomplete not filling gaps with new biomass #160

Open aabest opened 12 years ago

aabest commented 12 years ago

We have been editing the biomass file for the S. oneidensis 211586.9.906 model to reflect the published version of the model, but when we attempt to gapfill with this as the new BOF on complete media, we are getting problems. There are no error message, just a notification that autocompletion was successful. The model fails to grow on a check growth, however. See the message below:

The autocompletion process for S. oneidensis (211586.9) is not functioning properly. However, the autocompletion process for other Shewanella (S. violacea, S. woodyi, S. amazonensis, S. pealeana) seems to be working fine (we have not tried to change the automatically generated biomass for these models) . The main issue is that when mdl-autocomplete is run for S. oneidensis, the command executes, but no gap-filling reactions are added.

We've done some comparison of files produced in both successful and unsuccessful autocomplete cases. Here are some observations:

  1. Under CompleteGapfillingOutput.txt there should not be a biomass reaction listing with a “FAILED” next to it. There should also be a list of reactions with a ‘+’ before each listed reaction, and a 1/1 at the end of this line.
  2. When looking at the files made before and after auto-completion, there are no “AUTOCOMPLETION” reactions present for the new S. oneidensis model file (ending in ".tbl") in the second job directory. “AUTOCOMPLETION” reactions are present for the other four organisms.
  3. There is also a difference between the syntax of the biomass reactions in the model file, with that of S. oneidensis being: bio95000;=>;c;BOF;NONE;5;;
    Compared to the others: bio12895;=>;c;BOF;NONE;1;Biomass objective function;NONE and bio12905;=>;c;BOF;NONE;1;Biomass objective function;NONE

Here is a list of the commands that were run, in order, before examining the model files:

mdl-loadmodel -model Seed211586.9.906 -genome 211586.9 -biomassFile bio95000.bof -overwrite 1

fba-checkgrowth Seed211586.9.906 # there was no growth after this command

mdl-autocomplete Seed211586.9.906?Complete?1 # output said autocompletion was successful

fba-checkgrowth Seed211586.9.906 # still no growth occurred; no gaps were filled

We did run a mdl-loadbiomass command to get the biomass into the database prior to the above commands.

Thanks for looking into this... is there any other information that you need us to send?

samseaver commented 12 years ago

ModelSEED is limited in its potential by its database of balanced reactions. It is still possible for a biomass component to not be generated even if the entire database was at the organism's disposal (as is the case via gap-filling).

The gap-filling itself did not add any reactions. It completed without any errors, so it completed successfully, which is misleading, but, as you already know, the "FAILED" keyword means it could not grow all of the biomass components.

Can the S. oneidensis model be gap-filled with the biomass for the other Shewanella? If so, that's your starting point, and you'll have to iteratively investigate which extra biomass components are causing the problem. If you know which ones and why (incomplete pathway, missing drains) then you can add reactions via:

./bc-addbiochemistry

or drains via:

./mdl-changedrains

On Tue, Jun 12, 2012 at 7:30 AM, aabest reply@reply.github.com wrote:

We have been editing the biomass file for the S. oneidensis 211586.9.906 model to reflect the published version of the model, but when we attempt to gapfill with this as the new BOF on complete media, we are getting problems. There are no error message, just a notification that autocompletion was successful. The model fails to grow on a check growth, however. See the message below:

The autocompletion process for S. oneidensis (211586.9) is not functioning properly.  However, the autocompletion process for other Shewanella (S. violacea, S. woodyi, S. amazonensis, S. pealeana) seems to be working fine (we have not tried to change the automatically generated biomass for these models) .  The main issue is that when mdl-autocomplete is run for S. oneidensis, the command executes, but no gap-filling reactions are added.

We've done some comparison of files produced in both successful and unsuccessful autocomplete cases. Here are some observations:

  1. Under CompleteGapfillingOutput.txt there should not be a biomass reaction listing with a “FAILED” next to it. There should also be a list of reactions with a ‘+’ before each listed reaction, and a 1/1 at the end of this line.
  2. When looking at the files made before and after auto-completion, there are no “AUTOCOMPLETION” reactions present for the new S. oneidensis model file (ending in ".tbl") in the second job directory.  “AUTOCOMPLETION” reactions are present for the other four organisms.
  3. There is also a difference between the syntax of the biomass reactions in the model file, with that of S. oneidensis being: bio95000;=>;c;BOF;NONE;5;; Compared to the others: bio12895;=>;c;BOF;NONE;1;Biomass objective function;NONE  and bio12905;=>;c;BOF;NONE;1;Biomass objective function;NONE

Here is a list of the commands that were run, in order, before examining the model files:

mdl-loadmodel -model Seed211586.9.906 -genome 211586.9 -biomassFile bio95000.bof -overwrite 1

fba-checkgrowth Seed211586.9.906  # there was no growth after this command

mdl-autocomplete Seed211586.9.906?Complete?1  # output said autocompletion was successful

fba-checkgrowth Seed211586.9.906  # still no growth occurred; no gaps were filled

We did run a mdl-loadbiomass command to get the biomass into the database prior to the above commands.

Thanks for looking into this... is there any other information that you need us to send?

Reply to this email directly or view it on GitHub: https://github.com/ModelSEED/Model-SEED-core/issues/160

Postdoctoral Fellow Mathematics and Computer Science Division Argonne National Laboratory 9700 S. Cass Avenue Argonne, IL 60439

http://www.linkedin.com/pub/sam-seaver/0/412/168 samseaver@gmail.com (773) 796-7144

"We shall not cease from exploration And the end of all our exploring Will be to arrive where we started And know the place for the first time."    --T. S. Eliot