Multi-strain Modeling of Pseudomonas putida Strains for Metabolic Diversity Discovery

Joshua Mueller Author

04/02/2021 Added

42 Plays

Description

Using Genome Scale Models to explore the diversity of Pseudomonas putida strains.

Searchable Transcript

Search:

[00:00:01.080]Hello. I am Joshua Mueller,
[00:00:03.060]a graduate student in the chemical and biomolecular engineering department being
[00:00:07.260]advised by Dr. Wei Niu.
[00:00:09.390]Today I'm going to present my research on multi-strain modeling of pseudomonas
[00:00:13.050]putida strains for metabolic diversity discovery.
[00:00:17.820]Currently our economy is largely dependent on petroleum to create a broad number
[00:00:22.290]of consumer products ranging from fuels to plastics, to pharmaceuticals.
[00:00:27.210]One of the best alternatives to petrochemicals is using microorganisms for the
[00:00:31.590]production of these value added products from renewable resources.
[00:00:35.970]Pseudomonas putida has been shown to be a strong candidate for this process
[00:00:39.960]due to a number of its beneficial characteristics,
[00:00:42.630]such as being amenable to genetic manipulations,
[00:00:45.660]and also having a broad metabolism and tolerance to some industrial solvents.
[00:00:50.550]My research aims to help explore the diversity among strains of pseudomonas
[00:00:54.390]putida and in the process gain more understanding of the genetics that
[00:00:58.290]gives them their broad metabolic capabilities.
[00:01:02.550]To do this I utilize genome scale models.
[00:01:05.400]These are way to convert the metabolic reactions of an organism into a
[00:01:09.000]mathematical matrix.
[00:01:10.980]The computerized matrix then serves as the basis to explore how the reactions
[00:01:14.700]in genes are related to each other.
[00:01:17.130]Additional information can be incorporated into the models that provide info
[00:01:20.760]about the underlying genes necessary for reactions.
[00:01:24.660]Genome scale models can then be used for a variety of purposes,
[00:01:28.050]such as exploring the effects of gene knockouts or growth abilities in different
[00:01:31.890]environments.
[00:01:35.040]I used the well curated genome scale model of pseudomonas putida,
[00:01:38.340]KT2440 as a template to create strain specific models of a large
[00:01:42.990]number of previously uncharacterized strains.
[00:01:46.680]By identifying the homologous genes between strains,
[00:01:49.290]it is possible to create the framework of new models while skipping much of the
[00:01:52.920]more time intensive process that is typically involved in the construction of
[00:01:56.550]most genome scale models.
[00:01:59.920]For my project, 44 new strains of pseudomonas putida that were purchased from the American type
[00:02:05.310]culture collection and were fully sequenced by some of our collaborators
[00:02:10.140]To gather experimental data to validate and expand the models,
[00:02:13.380]the strains were grown on Biolog plates. As shown, Biolog plates are commercially
[00:02:18.330]available 96 well plates, where each well contains a different growth environment.
[00:02:23.580]Use of the plates and an appropriate plate reader then allows for testing of a
[00:02:27.150]wide variety of growth environments in a very short period of time.
[00:02:31.290]Our testing used biolog plates that contain different carbon sources,
[00:02:35.040]which allowed rapid evaluation of the strains growth capabilities on a variety
[00:02:38.940]of metabolites.
[00:02:42.270]The genomic and biolog data were then used to create a pan pseudomonas putida
[00:02:46.480]model that contained all of the genes and reactions that were present in any of
[00:02:50.280]the strains. As you can see,
[00:02:53.340]this model greatly expanded on the reference model of KT 2440 and added more
[00:02:57.960]than a hundred genes and over 150 new metabolic reactions.
[00:03:02.800]The biolog data was also used to test the ability for the individualized strain
[00:03:07.120]models to correctly predict growth on each of the carbon sources.
[00:03:11.620]The shown figure provides a summary for how well the models are at correctly,
[00:03:15.070]predicting growth on different carbon sources.
[00:03:18.490]Each column in the figure represents the predictions of a model for a specific
[00:03:22.120]strain of pseudomonas putida,
[00:03:23.920]and each row shows the predictions for a given metabolite.
[00:03:27.610]As you can see in the figure,
[00:03:28.840]there's a large block of true positives at the bottom that indicate that the
[00:03:32.500]models and experimental data are in agreement about the ability for growth on a
[00:03:36.130]given metabolite.
[00:03:37.900]There's also a large block in the middle of true negatives or the model and data
[00:03:41.560]are in agreement about the inability for growth.
[00:03:44.950]The false negatives represent areas where the model likely is lacking a
[00:03:48.220]metabolic reaction.
[00:03:49.390]And so doesn't predict growth when the biolog data indicates that it should.
[00:03:55.000]Those false negatives represent the first and easiest direction for future
[00:03:58.420]research by pointing to areas that we likely don't know enough about pseudomonas
[00:04:02.260]putida's metabolism and the underlying genes.
[00:04:06.071]These models do however
[00:04:07.510]show that this method can be a useful way to better explore a species' diversity
[00:04:12.040]without requiring months or years of experimental lab research.
[00:04:16.390]The models also can be used with a variety of tools that have already been
[00:04:19.780]created for genome scale models.
[00:04:22.870]Methods exist that can predict the effects of gene knockouts,
[00:04:25.990]as well as predicting optimal genetic mutations for production of a desired
[00:04:29.800]compound. If these models are combined with those methods,
[00:04:33.520]they could indicate strains that are optimal for a desired purpose without need
[00:04:37.510]of extensive experimental testing.
[00:04:41.230]I hope you enjoyed and learned from my presentation today.
[00:04:44.320]I would like to first and foremost, thank my advisor, Dr.
[00:04:46.690]Wei Niu for all the help she has been. I would also like to thank doctors,
[00:04:50.500]John Monk and Adam Feist.
[00:04:52.240]They are two of our collaborators that have helped the most with the
[00:04:55.090]modeling aspect of this work. Thank you once again for your attention.
[00:04:58.690]And I hope you have a good rest of your day.

The screen size you are trying to search captions on is too small!

You can always jump over to MediaHub and check it out there.

Comments

0 Comments

Multi-strain Modeling of Pseudomonas putida Strains for Metabolic Diversity Discovery

Description

Searchable Transcript

Comments icon comment

Related Channels

Comments