Workshop in computational molecular biology, April 15, 1991--April 14, 1994 Page: 4 of 17
This article is part of the collection entitled: Office of Scientific & Technical Information Technical Reports and was provided to UNT Digital Library by the UNT Libraries Government Documents Department.
Extracted Text
The following text was automatically extracted from the image on this page using optical character recognition software:
Computational aspects of the protein folding problem.
F. Cohen, Dept. of Pharmaceutical Chemistry, UCSF.
The amino acid sequence of a protein uniquely determines its tertiary struc-
ture. Deciphering this relationship, the protein folding problem, has become
increasingly important to molecular biologists. DNA sequencing has become
routine. Sequencing the human genome will produce a flood of protein pri-
mary structure data. Methods for determining the three dimensional structure
of a protein cannot keep pace with genomic data. Computational procedures
for relating sequence to structure are required. Two semi-empirical approaches
to the folding problem have evolved: detailed energy calculations and the hi-
erarchic condensation model. Energy calculations which attempt to describe
the thermodynamic transition between unfolded and folded states have become
hopelessly lost in multiple minima along the way. The hierarchic condensation
model postulates that secondary structure is a useful computational intermedi-
ate in protein folding. I will discuss the hierarchic condensation model and the
value of simplified representations of protein chains.
Secondary structure prediction. Pauling anticipated the locally regular or-
ganization of polypeptides into a-helices and /-sheets. X-ray crystallographic
structure determination has confirmed that these periodic elements dominate
protein architecture. Early attempts at relating amino acid sequence to sec-
ondary structure achieved 65% accuracy relying on exclusively local effects. Re-
finements of these methods have failed to improve their accuracy. Presumably,
this is because a significant fraction of the local peptide conformations are influ-
enced by non-local effects. We have exploited the non-local aspects of globular
structure to improve secondary structure prediction. Initial work focused on
loops, irregular segments of the chain which join secondary structure elements.
Consistent with their location on protein surfaces, loops (turns) are dominated
by hydrophilic residues. This tendency can be exploited to locate 'strong' loops.
Sequence pattern descriptors are written which embody this concept. 'Weak'
or context dependent loops can be located based on hydrophilicity and their se-
quential juxtaposition to 'strong' loops. By placing an upper limit on the length
of chain between successive loops, globularity is ensured and predictive accuracy
improves to about 90%. This improvement is predicated on an ability to recog-
nize whether the protein is all a-helical (a/a), all #-sheet (J3/3) or alternating
helix and sheet (a/3). We are extending this work to locate secondary structure
between loops. Preliminary work suggests that in a/a proteins, it is possible to
recognize the core of helices 89% of the time in a development set of 8 proteins
and 84% of the time in a control set of 9 proteins. Patterns to recognize the N
and C terminal caps of a-helices are currently under development.
Tertiary structure prediction. Our past computational experiments indicate
that secondary structure is a useful intermediate for predicting tertiary struc-
ture. Myoglobin is an instructive example. Six of the helices in myoglobin con-
tain clusters of hydrophilic residues along their surface. These docking sites can3
Upcoming Pages
Here’s what’s next.
Search Inside
This article can be searched. Note: Results may vary based on the legibility of text within the document.
Tools / Downloads
Get a copy of this page or view the extracted text.
Citing and Sharing
Basic information for referencing this web page. We also provide extended guidance on usage rights, references, copying or embedding.
Reference the current page of this Article.
Tavare, S. Workshop in computational molecular biology, April 15, 1991--April 14, 1994, article, April 12, 1995; United States. (https://digital.library.unt.edu/ark:/67531/metadc786310/m1/4/: accessed April 25, 2024), University of North Texas Libraries, UNT Digital Library, https://digital.library.unt.edu; crediting UNT Libraries Government Documents Department.