
               HOMOLOGOUS SEGMENT REPLACEMENT

    It  is sometimes  convenient to replace the atomic coor-
dinates of a segment of one 3D model with those of a homolo-
gous  segment of another 3D model. As an example, suppose we
are working with the 3D structure of model B and we want  to
take  advantage  of  the  known  3D  structure of one of its
branches, say a stem-loop having an NMR-determined structure
that  is  part  or all of a PDB file. We import the PDB file
into model A, define the segment of  interest  by  selecting
its  endpoints, define the segment of model B to be replaced
by selecting its endpoints, and then invoke the  replacement
algorithm.

    The  term  ’homologous’ is here used in the strict sense
that the corresponding segments have the same length and the
same  nucleotide  sequence. A further requirement is imposed
by how the replacement algorithm works. Assume the PDB  file
nucleotide  sequence  consists  of three parts C1, S1 and D1
such that C1-S1-D1 is the whole sequence and  S1 is the seg-
ment  of  interest.  Similarly, that C2-S2-D2 is the B model
sequence and S2 is the segment to be replaced by S1.   Coor-
dinates of the S1 nucs are first defined in terms of the ab-
solute coordinates of it first nucelotide, and the  relative
coordinates  of D2 nucs are defined in terms of the absolute
coordinates of the last nucleotide of S2.   The  replacement
is  then  implemented by the following steps: (1) assign the
first nucleotide of S1 the absolute coordinates of the first
nucleotide  of S2; (2) using the relative coordinates of S1,
assign to S2 absolute coordinates based on the new  absolute
coorindates of the first nucleotide of S1; (3) using the new
absolute coordinates of the last nucleotide  of  S2,  assign
new  absolute coordinates to the D2 segment. Segment S2 thus
acquires the conformation of segment S1 and the  D2  segment
is moved to accommodate this change while retaining its con-
formation.

                          The END






