How to select loops in protein structure?

Question

[qas]

Is it possible to automatically select loops in the protein or do i have to do it manually by typing residues numbers?

[lieven.sterck]

what do you consider as 'loops' ? and did you already identified them or is that part of the question as well?

[qas]

Loops are small fragments that link secondary structures in protein. I need to select loops from the protein structure - I don't know which residue is a loop.

[Joe]

In what tool? You can do this in UCSF chimera (at least in principle).

Answer 1

[Mensur Dlakic]

There are many programs that assign secondary structure from PDB files. For example, this is what STRIDE assigns to PDB structure 1ako:

REM  -------------------- Secondary structure summary -------------------  1AKO
REM                                                                        1AKO
CHN  1ako.pdb A                                                            1AKO
REM                                                                        1AKO
REM                .         .         .         .         .               1AKO
SEQ  1    MKFVSFNINGLRARPHQLEAIVEKHQPDVIGLQETKVHDDMFPLEEVAKL   50          1AKO
STR        EEEEEETTTGGG HHHHHHHHHHH TTEEEEETT   GGGTTHHHHHHH               1AKO
REM                                                                        1AKO
REM                .         .         .         .         .               1AKO
SEQ  51   GYNVFYHGQKGHYGVALLTKETPIAVRRGFPGDDEEAQRRIIMAEIPSLL  100          1AKO
STR         EEEEEEETTEEEEEEEETTTTTEEEE TTTT HHHHH EEEEEEEETT               1AKO
REM                                                                        1AKO
REM                .         .         .         .         .               1AKO
SEQ  101  GNVTVINGYFPQGESRDHPIKFPAKAQFYQNLQNYLETELKRDNPVLIMG  150          1AKO
STR       EEEEEEEEE   TTBTTT HHHHHHHHHHHHHHHHHHHH TTTT EEEEE               1AKO
REM                                                                        1AKO
REM                .         .         .         .         .               1AKO
SEQ  151  DMNISPTDLDIGIGEENRKRWLRTGKCSFLPEEREWMDRLMSWGLVDTFR  200          1AKO
STR       E      GGGB   HHHHHHHHHH BTTTTHHHHHHHHHHHHH EEEHHH               1AKO
REM                                                                        1AKO
REM                .         .         .         .         .               1AKO
SEQ  201  HANPQTADRFSWFDYRSKGFDDNRGLRIDLLLASQPLAECCVETGIDYEI  250          1AKO
STR       HHTTTTTTT  EETTTTTGGGG  EE  EEEEEEHHHHH EEEEEE HHH               1AKO
REM                                                                        1AKO
REM                .                                                       1AKO
SEQ  251  RSMEKPSDHAPVWATFRR                                  268          1AKO
STR       HHTTTT  B  EEEEE                                                 1AKO
REM                                                                        1AKO
REM                                                                        1AKO
REM                                                                        1AKO

Any letter (or empty space) under the sequence that is not E, G or H is a loop. But I am guessing that is not what you want. If you are using a particular program and need to identify loops without typing the residues manually, first you need to tell us what that program is.

[qas]

The program is pyRosetta.

[Mensur Dlakic]

Don't know how to do it in that program, but there is an extensive manual for it that likely covers residue selection based on their secondary structure.