Advanced

Testing Direct Coupling Analysis on HP Model Proteins

Kadhim, Mustafa LU (2018) FYTK02 20172
Computational Biology and Biological Physics
Department of Astronomy and Theoretical Physics
Abstract
Direct coupling analysis (DCA) models correlations in sets of related (homologous) protein sequences using a Potts-like spin model ansatz. From the couplings of the Potts model, derived by inverse statistical mechanics, residue-pair contacts in the 3D structure of the protein are predicted. In this thesis, this approach is applied to structures from the HP model on a square lattice. All HP sequences folding to the structures studied are known from previous work. For the calculation of the couplings, a maximum likelihood procedure is implemented, based on gradient descent and Monte Carlo methods.
Popular Abstract
For decades, we have been trying to understand the DNA, RNA and the proteins responsible for several functions in our bodies and organisms. Functions such as, reproduction of cells, transmitting information and transporting of molecules.

Since the DNA and RNA chains can be quite complex and long, we find it easier to deal with protein chains. Also, we know that proteins can help in binding and forming the DNA helix chains, thus it is sufficient to understand the behaviour of proteins and what provide their functions. After further research into the subject, it has been discovered that a protein’s structure plays an important role in deciding its function and where to operate inside a body.

A typical protein chain consists of a... (More)
For decades, we have been trying to understand the DNA, RNA and the proteins responsible for several functions in our bodies and organisms. Functions such as, reproduction of cells, transmitting information and transporting of molecules.

Since the DNA and RNA chains can be quite complex and long, we find it easier to deal with protein chains. Also, we know that proteins can help in binding and forming the DNA helix chains, thus it is sufficient to understand the behaviour of proteins and what provide their functions. After further research into the subject, it has been discovered that a protein’s structure plays an important role in deciding its function and where to operate inside a body.

A typical protein chain consists of a large number of organic molecules, amino acids, sitting next to each other, as a necklace. Each amino acid is interacting with other amino acids in the chain and sometimes with other nearby protein chains, depending on the circumstances. These types of interactions can develop into a complicated mathematical problem, that needs to take many variables into account. Also, the complexity of the calculations grows exponentially with the number of amino acids in the chain. Scientists have been trying to solve the problem of predicting a protein's structure by knowing its amino acid components. Many methods have been tried and proposed. One of these methods requires taking advantage of inverse statistical procedures such as a direct coupling analysis (DCA) methodology, where it is considered useful to reveal contacts between amino acids in the protein sequence, in which later on can be used to predict the structure of the protein. The DCA method compares many protein sequences with each other, aiming to find a correlation between the amino acids that influence the protein’s appearance (shape). From this method, it turns out that one often can infer pairs of amino acids in contact. Knowledge of such contacts can greatly facilitate structure prediction.

If we succeed to predict the structure of proteins, only by knowing their amino acid sequences, we can start designing our own proteins, able to function as we desire at the required location in the body. Furthermore, we will be able to understand more about cancer, tumours and neurodegenerative diseases.

In conclusion, determining protein shapes can have a huge beneficial impact on our lives, from increasing our life expectancy to answering fundamental questions in biology. Thus, it is an important problem that we need to solve. (Less)
Please use this url to cite or link to this publication:
author
Kadhim, Mustafa LU
supervisor
organization
course
FYTK02 20172
year
type
M2 - Bachelor Degree
subject
keywords
Direct Coupling Analysis HP model proteins
language
English
id
8934116
date added to LUP
2018-02-07 14:16:50
date last changed
2018-06-12 11:16:22
@misc{8934116,
  abstract     = {Direct coupling analysis (DCA) models correlations in sets of related (homologous) protein sequences using a Potts-like spin model ansatz. From the couplings of the Potts model, derived by inverse statistical mechanics, residue-pair contacts in the 3D structure of the protein are predicted. In this thesis, this approach is applied to structures from the HP model on a square lattice. All HP sequences folding to the structures studied are known from previous work. For the calculation of the couplings, a maximum likelihood procedure is implemented, based on gradient descent and Monte Carlo methods.},
  author       = {Kadhim, Mustafa},
  keyword      = {Direct Coupling Analysis HP model proteins},
  language     = {eng},
  note         = {Student Paper},
  title        = {Testing Direct Coupling Analysis on HP Model Proteins},
  year         = {2018},
}