Not Really The End: Results and Future Directions
Hello everyone, and welcome back to the last blog post!
I can not believe it has already been 10 weeks of working on this project. Somehow, here we are, so let’s go over the findings.
The major takeaway from my project is that there does seem to be a correlation between the classification of a bitter compound and its health outcome. The classification metric that I saw correlated was the superclass and class of each compound. Unfortunately, the bitterness intensity did not return any significant results.
To determine the significance of the results for the superclass and class, I ran a chi-squared test to see if there was any way the distribution of compounds was due to chance or if there was another variable at play. The p-value had to be less than 0.05 to be considered significant.
For the anti-cancer group, the most prolific superclass was benzenoids, with a p-value of 0.040699 after the chi-squared test was run. The only other superclass that even came slightly close to the number of benzenoids was the alkaloids and derivatives. This was interesting because although benzenoids are actually not that great for you normally and can even be linked to causing cancer, they are amazing at fighting cancer. Once additional studies are conducted confirming the efficacy of benzenoids over other bitter compounds in cancer treatment, future studies could work towards developing targeted treatments using benzenoids.
Next, for the anti-inflammatory group, the most prevalent superclass was phenylpropanoids and polyketides. The p-value for this group was 0.001457. The 2 other superclasses that were closest despite being off by a bit were the organoheterocyclic compounds and the alkaloids and derivatives. I wanted to perform statistical analysis on the class data, but unfortunately, the expected counts were not met. However, just looking at the distribution, flavonoids by far were the most prevalent class.
In the enteroendocrine superclass distribution, phenylpropanoids and polyketides, alkaloids and derivatives, and organoheterocyclic compounds were all similar in levels of prevalence. The p-value for this group was 0.0034341. When looking at the class distribution from these 3 superclasses, the most prolific class was again flavonoids. The p-value for this was 0.028906.
Finally, for the respiratory group, the largest two superclasses were alkaloids and derivatives and organic acids and derivatives. The p-value for this distribution was 0.01293111. When looking into the class distribution for this group, the most prevalent classes were quinolines and derivatives, cinchona alkaloids, and carboxylic acids and derivatives. The p-value for this distribution was 0.0269834.
Bitter compounds in the superclasses of lipids and lipid-like compounds, hydrocarbons, organic oxygen compounds, and benzenoids were the least prevalent in every single category, with the exception of benzenoids for the anti-cancer group. These results line up with my hypothesis that not all bitter is the same. There are superclasses of bitter that are likely more effective for certain health conditions.
However, this systematic review is not the end. Further research needs to be done to confirm the results. I would recommend that the superclasses in each category be tested alongside each other in one study to ascertain their effectiveness. This will help solidfy the results seen in this study if the effectiveness lines up with the most prevalent superclasses. Then, researchers can work towards making supplements, treatments, or diet plans that help with these conditions since we have seen how powerful bitterness can be.