A high resolution X-ray crystallography experiment can give "a set of free atoms that reproduce the density well" (Morris et al. 2002). The set of free atoms can be represented by a set of 3-D coordinates, which are the centres of the heavy atoms (the carbon, nitrogen, oxygen and sulphur atoms) in a protein, but the identities of the atoms are initially unknown (i.e. we don't initially know the element type of each atom, and we don't know which residue each atom belongs to).
A useful first step in interpreting the free atoms is to try to identify a subset of the free atoms that can be assigned to be the alpha-carbons in a protein's main chain. In doing this, we can use the fact that the centres of consecutive alpha-carbon atoms are about 3.8 Å apart.
Write programs that can read the coordinates of a set of free atoms from a file, and then identify the order of alpha-carbon atoms in the chain.
Start with a simple case, where all of the free atoms in the input file are alpha-carbon atoms, and we just need to figure out their order in the chain. The direction of the chain doesn't matter (we have no way of knowing whether the chain of alpha-carbons that we find is in the direction from the N-terminal to the C-terminal, or in the reverse direction).
Test your program with file data_q1.txt. Each line of that file contains an atom serial number followed by the x-, y- and z-coordinates of a free atom.
Now consider a data file where only some of the free atoms in the input file are alpha-carbon atoms. Find the longest chain of candidate alpha-carbon atoms. For this exercise, the direction of the chain doesn't matter (that is, it doesn't mater if the candidate alpha-carbon atoms in the output are listed in the direction from the N-terminal to the C-terminal, or in the reverse direction).
Test your program with file data_q2.txt. Each line of that file contains an atom serial number followed by the x-, y- and z-coordinates of a free atom.
(The program described by Morris et al. does much more than your are expected to do in this exercise. You might get some inspiration from looking at their paper, but you are not expected to implement everything that they did!)
Q.
Can we have some more test data, with sample output?
[2018-11-27]
A.
Running your program for question 1 on file
test_q1.txt
should give the following output (or the same sequence in reverse):
8 10 9 7 6 5 3 1 4 2
Q.
How much can the distance between consecutive alpha-carbon atoms vary from
3.8 Å ? [2018-11-27]
A.
You can choose your own value for this.
As a guide, you could modify slightly your program from Practical 2
Question 1 to print out values for distances between consecutive alpha-carbon
atoms, or their difference from 3.8 Å, to see what values are typical.
Q.
I have found a chain of over 60 candidate alpha-carbon atoms - is that OK?
[2018-11-30]
A.
The protein used to generate the test data file has between 40 and 50 residues.
It is likely that your program is working "correctly", and is finding a
feasible chain of free atoms that is even longer than the chain of "true"
alpha-carbon atoms.
This could give a good grade, but it could be possible to improve the
solution further.
In the referenced paper, the authors look for other main chain atoms in the
peptide plane between two candidate alpha-carbons (Figure 2).
Another approach would be to consider the "pseudo-valence angles" defined
by three consecutive alpha-carbon atoms (Section 2.5 of the referenced paper).
As a guide, you could modify slightly your program from Practical 2
Question 1 to print out values for angles defined by three consecutive
alpha-carbon atoms to see what values are typical.
You should submit your solutions via the Fire system before 23:59 on Wednesday 5 December 2018.
Everyone should attempt question 1. Those aiming for a higher grade should also attempt question 2.
Upload the following to the Fire system:
Your programs for each of the questions that you attempt, together with instructions on how to compile and run your programs.
A brief description of the approach taken to solve the problems. State any assumptions that you make about the data.
A file that shows the result of compiling and running your programs. For each question, the output should contain a list of the serial numbers of the atoms that make up the identified chain of alpha-carbons. Your program should write one atom serial number per line. it should also write out the total number of alpha-carbon atoms in the chain.
Ensure that your name is included in all files submitted (e.g. in a comment in your source code).