Difference between revisions of "Team:NCTU Formosa/Dry Lab/Microbiota Prediciton"

Line 322: Line 322:
 
       <img src="https://static.igem.org/mediawiki/2018/3/37/T--NCTU_Formosa--real.png" class="real">
 
       <img src="https://static.igem.org/mediawiki/2018/3/37/T--NCTU_Formosa--real.png" class="real">
 
       <img src="https://static.igem.org/mediawiki/2018/1/11/T--NCTU_Formosa--Predictive.png" class="predictive">
 
       <img src="https://static.igem.org/mediawiki/2018/1/11/T--NCTU_Formosa--Predictive.png" class="predictive">
 
+
      <img src="https://static.igem.org/mediawiki/2018/4/4e/T--NCTU_Formosa--weka_OTU_table.jpg"class="July&predictive July bacteria ratio">
 
     </div>
 
     </div>
 
     <div class="title_1"><p>References</p></div>
 
     <div class="title_1"><p>References</p></div>

Revision as of 00:34, 18 October 2018

Navigation Bar Microbiota Prediction

     Machine learning allows us to realize the seemingly insurmountable goal of predicting the fluctuations of entire microbiotas due to the specific effects of bio-stimulators. While traditional ecologists may find it too difficult to consider every unique microbial relationship in an ecosystem, machine learning programs uses numerical analysis to not only quickly determine these associations but also use them to predict overall population shifts caused by stimuli. For our modelling purposes we choose Weka, a software with strong classification capabilities, to establish accurate connections between every genera of bacteria in our soil.

Construction

     Weka predicts overall shifts in microbiota by determining how a target genus is altered, then using correlation values between the target genus and other genera to calculate how the rest of the bacteria change.

     This is where Weka’s support of classification is useful to us, as it can quickly determine which genera affect each other and which do not. We begin with our analyzed NGS data – specifically, a heat map detailing correlation values of bacteria in our soil sample. Weka allows us to set our own conditions to detect correlation between two genera of bacteria and will filter the results to yield all the pairs of bacteria determined to be correlated, either positively or negatively. Here, we take June's correlation heatmap of the top 20 bacteria as an example:

Figure 1: Temperature growth curve model progress

     Next, We define positive correlation as any Spearman's Correlation value above 0.7, and negative correlation as any value below -0.7. For example:

     Once we’ve determined the significantly correlated pairs of bacteria, Weka can further establish the exact nature of these relationships. Not all bacterial relationships exhibit linear regression, so Weka then takes the correlated pairs and plots the data of each pair into a graph to determine the true nature of each association.

$$y=-4.1248x^2+0.2061x+0.0008;\ R^2=0.725$$

Figure 2: Example of positive non-linear correlation

$$y=-0.061x+0.0003;\ R^2=0.714$$

Figure 3: Example of negative linear correlation

Results

References

1. Bouckaert, R. R., et al. (2013). "WEKA Manual for Version 3-7-8, 2013." 21.

2. WI, H., et al. (2011). "Practical machine learning tools and techniques."

3. Barabasz, W. and J. J. P. J. o. E. S. Lipiec (2002). "Biological effects of mineral nitrogen fertilization on soil microorganisms." 11(3): 193-198.

4. KUMAR, A. and L. C. J. P. RAI (2017). "Soil Organic Carbon and Availability of Soil Phosphorus Regulate Abundance of Culturable Phosphate Solubilizing Bacteria in Paddy Fields of the Indo-Gangetic Plain."

5. Lambert, R. J. J. J. o. a. m. (2011). "A new model for the effect of pH on microbial growth: An extension of the Gamma hypothesis." 110(1): 61-68.

6. Nihala Jabin, P. (2017). Screening of potash solubilizing bacteria for plant growth promotional activity and nutrient uptake of brinjal, Vasantrao Naik Marathwada Krishi Vidyapeeth, Parbhani.

7. Ratkowsky, D. A., et al. (1983). "Model for bacterial culture growth rate throughout the entire biokinetic temperature range." J Bacteriol 154(3): 1222-1226.

8. Rousk, J., et al. (2011). "Bacterial salt tolerance is unrelated to soil salinity across an arid agroecosystem salinity gradient." 43(9): 1881-1887.

9. Wikipedia contributors. (2018, October 12). Bacillus. In Wikipedia, The Free Encyclopedia. Retrieved 18:17, October 16, 2018, from https://en.wikipedia.org/w/index.php?title=Bacillus&oldid=863696818

10. Wikipedia contributors. (2018, March 23). Geobacter. In Wikipedia, The Free Encyclopedia. Retrieved 18:34, October 16, 2018, from https://en.wikipedia.org/w/index.php?title=Geobacter&oldid=831990733

11. Espenberg, M., et al. (2018). "Differences in microbial community structure and nitrogen cycling in natural and drained tropical peatland soils." Scientific Reports 8(1): 4742.

12. Hou, J., et al. (2015). "PGPR enhanced phytoremediation of petroleum contaminated soil and rhizosphere microbial community response." Chemosphere 138: 592-598.

13. Hruska, K., Vyzkumny Ustav Veterinarniho Lekarstvi, Brno (Czech Republic) and M. Kaevska, Vyzkumny Ustav Veterinarniho Lekarstvi, Brno (Czech Republic) (dec2012). "Mycobacteria in water, soil, plants and air: a review." v. 57.

14. Jiao, S., et al. (2016). "Microbial succession in response to pollutants in batch-enrichment culture." Scientific Reports 6: 21791.

15. Leys, N. M. E. J., et al. (2004). "Occurrence and Phylogenetic Diversity of Sphingomonas Strains in Soils Contaminated with Polycyclic Aromatic Hydrocarbons." Applied and Environmental Microbiology 70(4): 1944-1955.

16. Ma, M., et al. (2018). "Effect of long-term fertilization strategies on bacterial community composition in a 35-year field experiment of Chinese Mollisols." AMB Express 8(1): 20.

17. Martineau, C., et al. (2015). "Comparative analysis of denitrifying activity in Hyphomicrobium nitrativorans, Hyphomicrobium denitrificans and Hyphomicrobium zavarzinii." AEM. 00848-00815.

18. Rodgers-Vieira, E. A., et al. (2015). "Identification of Anthraquinone-Degrading Bacteria in Soil Contaminated with Polycyclic Aromatic Hydrocarbons." Applied and Environmental Microbiology.

19. Sangwan, P., et al. (2005). "Detection and cultivation of soil verrucomicrobia." Appl Environ Microbiol 71(12): 8402-8410.

20. Sorensen, J. and O. Nybroe (2004). Pseudomonas in the Soil Environment. Pseudomonas: Volume 1 Genomics, Life Style and Molecular Architecture. J.-L. Ramos. Boston, MA, Springer US: 369-401.

21. Umadevi, P., et al. (2018). "Trichoderma harzianum MTCC 5179 impacts the population and functional dynamics of microbial community in the rhizosphere of black pepper (Piper nigrum L.)." Brazilian Journal of Microbiology 49(3): 463-470.

22. van Dijl, J. M. and M. Hecker (2013). "Bacillus subtilis: from soil bacterium to super-secreting cell factory." Microb Cell Fact 12: 3.

23. Wang, R., et al. (2017). "Microbial community composition is related to soil biological and chemical properties and bacterial wilt outbreak." Scientific Reports 7(1): 343.

24. Winston, M. E., et al. (2014). "Understanding Cultivar-Specificity and Soil Determinants of the Cannabis Microbiome." PLOS ONE 9(6): e99641.

25. Yan, G., et al. (2017). "Effects of different nitrogen additions on soil microbial communities in different seasons in a boreal forest." 8(7): e01879.

Template