Analysing shifts in protein structure in relation to drug resistance in Mycobacterium tuberculosis

Analysing shifts in protein structure in relation to drug resistance in Mycobacterium tuberculosis

Marinus, tycho (TU Delft Electrical Engineering, Mathematics and Computer Science; TU Delft Pattern Recognition and Bioinformatics)

Meehan, Conor (mentor)
Abeel, Thomas (mentor)

Delft University of Technology

2018-09-20

While the sequencing era has greatly increased the amount of protein sequences, in the eld of protein structure this increase has not occurred, mostly due to the high cost and workload of protein structure determination. This has resulted in a large gap between the amount of protein sequences and protein structures. In the case of TB drug resistance, this combined with the time consuming process of drug phenotype testing is posing a challenge for resolving the global health problem of drug resistant TB. Wehavebuilt SSP, a sequence to structure pipeline that aims to reduce the gap between sequence and structure availability using high throughput in silico protein structure prediction methods to produce protein structures. This is combined with a drug phenotype labeling based on phylogenetic properties of the genomic data. This combination attempts to tackle both the lack of protein structures and time consuming method for drug phenotype testing at the same time. We demonstrate the workings of SSP on two proteins of Mycobacterium tuberculosis. While we show that SSP creates protein structures that are consistent and reliable, the genotype-based labeling does not perform well for predicting drug resistant conveying mutations.

Tuberculosis
Protein folding
Prediction

http://resolver.tudelft.nl/uuid:ab1dd507-fa29-457a-9adf-eb55a17f923b

Student theses

master thesis

© 2018 tycho Marinus