Comparing the efficacy of algorithmic approaches to the functional annotation of proteins in Trichomonad parasites
(2022) BINP52 20212Degree Projects in Bioinformatics
- Popular Abstract
- How useful are algorithms when studying the genetics of non-model organisms?
Metamonada is a group of eukaryotic single-celled organisms called protists that live in different environments using varied metabolic pathways. One unifying characteristic of the supergroup is their incredibly functionally reduced mitochondria. This group of organisms is worth studying because they are a mixture of free-living and parasitic organisms, among which parasitism has evolved independently multiple times. My main organism of interest is one of these parasites: Trichomonas vaginalis, the most common non-viral sexually transmitted infection in humans.
For my thesis project, I used a variety of computational approaches to analyze the proteins of... (More) - How useful are algorithms when studying the genetics of non-model organisms?
Metamonada is a group of eukaryotic single-celled organisms called protists that live in different environments using varied metabolic pathways. One unifying characteristic of the supergroup is their incredibly functionally reduced mitochondria. This group of organisms is worth studying because they are a mixture of free-living and parasitic organisms, among which parasitism has evolved independently multiple times. My main organism of interest is one of these parasites: Trichomonas vaginalis, the most common non-viral sexually transmitted infection in humans.
For my thesis project, I used a variety of computational approaches to analyze the proteins of these organisms. Functional annotation and targeting prediction tools use existing knowledge to try to predict protein functions and where proteins will function inside of a cell; and orthologous clustering identifies gene families. Looking at gene families is particularly important when studying metamonads, because these organisms are understudied and have complicated genomes, so the regular functional annotation and targeting prediction tools often don’t work. However, proteins within gene families often maintain similar functions. Looking at the differences in distribution of gene family members between the parasitic and free-living organisms can help us spot trends and learn about the evolution of parasitism within these species.
I was particularly interested in secreted and mitochondrial proteins because of the medical implications. Secreted proteins – which are found on the cell membrane or actively secreted from the cell – are the proteins that most directly interact with the host’s immune system. Since the mitochondria of these organisms are so reduced, they often function very differently from their hosts, and can therefore serve as viable drug targets.
Metamonads show the importance of experimental studies on non-model organisms.
I used proteins which had been experimentally determined to be targeted to the secretome or the mitochondria as a control dataset. It became clear upon examining the algorithmic predictions for these proteins that the programs do not accurately predict protein targeting among these organisms. This underscores the importance of working with non-model organisms in experimental settings – the algorithms are not enough.
My thesis can be described as a project where I primarily “did the same thing multiple times slightly differently,” and this enabled me to see both the limitations of algorithms and bioinformatics, as well as the importance of comparing results between multiple approaches, in order to produce the most robust results.
Master’s Degree Project in Bioinformatics 60 credits 2022
Department of Biology, Lund University
Advisor: Courtney Stairs
Microbiology Group, Department of Biology, Lund University (Less)
Please use this url to cite or link to this publication:
http://lup.lub.lu.se/student-papers/record/9102872
- author
- Varga, Virág
- supervisor
- organization
- course
- BINP52 20212
- year
- 2022
- type
- H2 - Master's Degree (Two Years)
- subject
- language
- English
- id
- 9102872
- date added to LUP
- 2022-11-07 11:14:33
- date last changed
- 2022-11-07 11:14:33
@misc{9102872, author = {{Varga, Virág}}, language = {{eng}}, note = {{Student Paper}}, title = {{Comparing the efficacy of algorithmic approaches to the functional annotation of proteins in Trichomonad parasites}}, year = {{2022}}, }