Skip to content

ACRMGroup/findcore

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

6 Commits
 
 
 
 
 
 
 
 
 
 

Repository files navigation

findcore

Andrew C.R. Martin, UCL, 1996-2025

All programs updated in November 2025 to use the new BiopLib library and findcora tidied up to match our coding standards.

findcore

The original findcore program was written in 1996 as part of the analysis of the CASP2 comparative (template-based) modelling assessment. This was based on taking two structures fitted using Christine Orengo's SSAP software which performed a structural alignment and contained secondary structure assignments in its output.

The program reads the SSAP alignment and performs a fit between the structures based on that alignment. It then throws out residue pairs that have a C-alpha separation of >3A and grows the fitted regions while the deviation C-alpha separation is <=3A. The fitting and zone adjustment then iterated until no more residue pairs are added or removed. (The default 3A cutoff can be changed on the command line.)

The final set of residues used for fitting is defined as the core.

findcora

findcora is a modification of the findcore program done by Gabby Marsden (nee Reeves), a PhD student working with Christine Orengo and Andre Martin. She modified the program so that it could work with multiple structure alignments from Cora rather than pairwise alignments from SSAP.

profitcore

ProFit is Andrew Martin's protein least-squares fitting software. After findcore was written, ProFit was enhanced to add an iterative structural alignment feature. This uses a seed alignment (typically the sequence alignment) as a proxy for the structural alignment; it then fits the structures and uses a dynamic programming algorithm to update the equivalenced residues, and at the same time brings in residue pairs with C-alpha separation of <=3A and throwing out those with separation >3A. This then iterates to convergence. As with findcore, this 3A cutoff can be changed.

Consequently, ProFit is doing a combination of what SSAP and findcore were doing.

profitcore is a new program (November 2025) that takes the sequentially numbered zones generated by ProFit and converts them to residue number ranges (using the residue number labels found in the PDB files). Optionally it will write PDB files where the zones used for fitting (i.e. the core) have a B-value of 1.0 while all other atoms have a B-value of 0.0.

Currently profitcore only handles 2 structures (i.e. like SSAP and findcore), but will soon be updated to handle multiple structure (like CORA and findcora).

Compilation

Compilation requires our bioplib library to be installed. It is then just a case of typing make in the src directory.

About

A program to find core regions of proteins using an approach suggested by Cyrus Chothia

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

 
 
 

Contributors

Languages