Asking questions about protein structure determination

In summary: However, there are cases where additional proteins (chaperone proteins) can interact the growing peptide chain, adding influences on the protein's stable configuration that are independent of the amino acid sequence's influence on the ultimate shape the protein settles into. In summary, there are two main approaches to protein structure prediction: homology modeling and physics-based modeling. Homology modeling compares the target protein's amino acid sequence to known protein structures, while physics-based modeling uses a force field to predict the protein's structure from scratch. Homology modeling is currently the most accurate and efficient method, but there have been significant advancements in physics-based
  • #1
docnet
Gold Member
690
343
TL;DR Summary
Asking questions about protein structure determination
Hi,

I tend to write down questions/thoughts that come to me as I read articles that are interestsing. I'm reading the intro to solution state NMR in a biophysics textbook.

Any there any purely theoretical approaches to determine protein structure / dynamics ?
The 1 dimensional digital code of DNA is easily obtained. And, wouldn’t it be more efficient to develop a sophisticated computer software that runs on pure computational power to determine the 3d structure, when we input the 1 dimensional data? Would this immense computational power? Does P = NP?

Leonard Susskind said calculations involving the folding of a hypothetical molecule of a hundred atoms results in an energy diagram with over 10^100 discrete “valleys” representing locally stable conformations. Still, with constraints given by the amino acids and their sequence, we would be able to lower that number significantly. Given proteins obey the laws of thermodynamics, will the task of determining protein structure be no longer a difficult task one day?
 
Biology news on Phys.org
  • #2
docnet said:
Any there any purely theoretical approaches to determine protein structure / dynamics ?
The 1 dimensional digital code of DNA is easily obtained. And, wouldn’t it be more efficient to develop a sophisticated computer software that runs on pure computational power to determine the 3d structure, when we input the 1 dimensional data? Would this immense computational power? Does P = NP?

In general, there are two approaches to protein structure prediction 1) homology modeling and 2) physics-based modeling.

Homology modeling makes the assumption that similar sequences will lead to similar structures. These approaches will compare the amino acid sequence of interest to the many thousands of known protein structures and uses that information to guide structure prediction. In contrast, physics-based models start with just the amino acid sequence and a "force field" describing the various intermolecular forces that govern protein folding and use that information to predict the protein's structure ab initio.

Homology modeling approaches are by far the most accurate and efficient methods available, and their power only grows as more protein structures are determined. Of course, the most effective modeling tools combine elements of both approaches. For example, they would use homology modeling to build an initial model for the protein then refine the model using physics based methods.

There have, however, been major improvement to physics-based methods in the past decade. For example, a former hedge fund manager interested in the protein folding problem built a supercomputer specifically for performing protein folding simulations and has had success modeling the folding of small, fast folding proteins (see for example, https://www.nature.com/news/2010/101014/full/news.2010.541.html).
 
  • Like
Likes epenguin, jim mcnamara, docnet and 1 other person
  • #3
Along the lines of the physics based approach, there are additional factors, beyond just the amino acid sequence, that can affect the 3D configuration.

Proteins are made from one end to the other.
As the forming amino acid string is extruded from the ribosome, the initial groups of amino acids will form little bits of more stable 3D configuration.
After later parts are added and extruded, new conformations might arise for the already extruded parts.
This might actually make the calculations easily due to the fewer parts involved initially.
Interactions with the parts of ribosomal RNA's and proteins that are exposed to the elongating peptide chain may also influence the conformations that the growing chain of amino acids experience as stable. Presumably the ribosomal influences would be common to most/all proteins, since most ribosomes would be expected to be the same.

However, there are cases where additional proteins (chaperone proteins) can interact the growing peptide chain, adding influences on the protein's stable configuration that are independent of the amino acid sequence's influence on the ultimate shape the protein settles into.

These can add additional wrinkles to making predictions.
 
  • Like
Likes docnet
  • #4
Ygggdrasil said:
In general, there are two approaches to protein structure prediction 1) homology modeling and 2) physics-based modeling.

Homology modeling makes the assumption that similar sequences will lead to similar structures. These approaches will compare the amino acid sequence of interest to the many thousands of known protein structures and uses that information to guide structure prediction. In contrast, physics-based models start with just the amino acid sequence and a "force field" describing the various intermolecular forces that govern protein folding and use that information to predict the protein's structure ab initio.

Homology modeling approaches are by far the most accurate and efficient methods available, and their power only grows as more protein structures are determined. Of course, the most effective modeling tools combine elements of both approaches. For example, they would use homology modeling to build an initial model for the protein then refine the model using physics based methods.

There have, however, been major improvement to physics-based methods in the past decade. For example, a former hedge fund manager interested in the protein folding problem built a supercomputer specifically for performing protein folding simulations and has had success modeling the folding of small, fast folding proteins (see for example, https://www.nature.com/news/2010/101014/full/news.2010.541.html).

Thank you for the detailed answer and the link. I enjoy reading your intelligent responses to PF on a regular basis.
 
  • #6
docnet said:
Thank you for the detailed answer and the link. I enjoy reading your intelligent responses to PF on a regular basis.

Thanks! Here are two articles on the subject of protein folding that have more information on predicting protein structure. The first is a shorter, more general and accessible review:
https://science.sciencemag.org/content/338/6110/1042.full

While the second is a much longer and more technical review article that goes into a lot more detail:
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC2735084/

Both are somewhat old articles as I have not followed the field so closely in recent years, so unfortunately, I don't know off the top of my head any good newer pieces that might be of interest.
 
  • Like
Likes docnet

What is protein structure determination?

Protein structure determination is the process of determining the three-dimensional arrangement of atoms in a protein molecule. This information is crucial for understanding the function of a protein and its interactions with other molecules.

What are the main methods used for protein structure determination?

The main methods used for protein structure determination include X-ray crystallography, nuclear magnetic resonance (NMR) spectroscopy, and cryo-electron microscopy (cryo-EM). Each method has its own advantages and limitations, and the choice of method depends on the size and complexity of the protein.

How does X-ray crystallography work for protein structure determination?

X-ray crystallography involves growing crystals of a protein and then exposing them to X-rays. The X-rays diffract off the atoms in the crystal, producing a diffraction pattern that can be used to determine the positions of the atoms in the protein. This method is particularly useful for larger proteins with well-ordered structures.

What is the role of NMR spectroscopy in protein structure determination?

NMR spectroscopy involves studying the interactions between the nuclei of atoms in a protein and a strong magnetic field. This method can provide information about the distances between atoms in a protein, which can be used to determine its structure. NMR spectroscopy is especially useful for studying proteins in solution.

What are the challenges in protein structure determination?

Protein structure determination can be a challenging process due to the complexity and variability of protein structures. Some proteins may be difficult to crystallize, and others may have multiple conformations or be highly flexible, making it difficult to obtain an accurate structure. Additionally, the interpretation of data from different methods can be complex and requires specialized expertise.

Similar threads

  • Biology and Medical
Replies
2
Views
13K
  • Beyond the Standard Models
Replies
24
Views
7K
  • Beyond the Standard Models
Replies
20
Views
7K
  • Special and General Relativity
3
Replies
94
Views
8K
  • MATLAB, Maple, Mathematica, LaTeX
Replies
5
Views
3K
  • MATLAB, Maple, Mathematica, LaTeX
Replies
2
Views
2K
  • MATLAB, Maple, Mathematica, LaTeX
Replies
7
Views
2K
  • MATLAB, Maple, Mathematica, LaTeX
Replies
9
Views
2K
Back
Top