No subject

viewed as sets of interacting molecules. To unravel their inner workings an=
d
to modulate their activities, a fundamental problem in computational biolog=
y
is probing mechanism and specific design of molecular interactions. In this
talk, I will address both aspects as optimization problems and explore new
insights into formulating and solving the problem.

The first part of the talk examines a molecule=92s ability to change its
conformation when binding to other molecules. This phenomenon, known as
conformational flexibility or molecular plasticity, has been hypothesized a=
s
a mechanism to facilitate tight and specific interactions. To test this
hypothesis and to incorporate it into molecular design, we extended charge
optimization theory to treat flexible molecules. Specifically, we framed an=
d
solved a single-objective optimization problem for binding affinity and a
corresponding multi-objective optimization problem for binding affinity and
specificity by adjusting the charge distribution of one binding partner to
be complementary for the other while permitting varying levels of
conformational flexibility. Application to a ligand that was developed to
inhibit HIV-1 protease identified ways to improve its binding affinity, and
indicated that a more favorable Pareto frontier of the binding
affinity=96specificity trade-off can be achieved with a more flexible ligan=
d.

The second part aims to understand molecular mechanisms by which small
molecules can exhibit binding promiscuity and to develop design strategies
to implement such promiscuity. We chose to study the inhibition of HIV-1
protease, which remains a tremendous challenge in the face of an evolving
viral population. Using computational design we constructed small-molecule
inhibitors targeting a set of wild-type and drug-resistant mutant HIV-1
proteases. Each design was solved as a combinatorial optimization problem i=
n
a discrete chemical and conformational space. Subsequent statistical
analysis revealed significant trends for promiscuous inhibitors: they tende=
d
to be smaller, more flexible, and more hydrophobic compared to highly
selective ones. Furthermore, structural analysis indicated that flexible
inhibitors often achieved promiscuity by conformational adaptations to
mutations in proteases. Our flexibility measure also showed its potential a=
s
a design criterion for promiscuous inhibitors because inhibitors with highe=
r
flexibility measures were more likely to be promiscuous. Finally, as no
inhibitor covered all variants, we designed small cocktails of inhibitors t=
o
do so through solution of a set cover problem.

This talk examines two perspectives on molecular complementarity through
analysis and design in an optimization framework.

--=20
Liv Leader
Faculty Services

Toyota Technological Institute
6045 S Kenwood Ave, #504
Chicago, IL 60637
Phone- (773) 834-2567
Fax-     (773) 834-9881
Email-  lleader at ttic.edu <jam at ttic.edu>
Web-   www.ttic.edu

--002215048b973981b2049dab66a7
Content-Type: text/html; charset=windows-1252
Content-Transfer-Encoding: quoted-printable

What:=A0=A0=A0=A0 <b>Tuesday, March 8 @ 11</b><br><br>Where:=A0=A0 <b>TTIC =
Conference Room #526</b>, 6045 S. Kenwood Ave, 5th Floor<br><br>Who:=A0=A0=
=A0=A0 <b>Yang Shen</b>, MIT<br><br>Title:=A0=A0=A0=A0=A0 <b>Biomolecular A=
nalysis, Design, and Engineering Through Optimization</b><br>

<br>From the perspective of networks, biological systems such as cells can =
be viewed as sets of interacting molecules. To unravel their inner workings=
 and to modulate their activities, a fundamental problem in computational b=
iology is probing mechanism and specific design of molecular interactions. =
In this talk, I will address both aspects as optimization problems and expl=
ore new insights into formulating and solving the problem.<br>

<br>The first part of the talk examines a molecule=92s ability to change it=
s conformation when binding to other molecules. This phenomenon, known as c=
onformational flexibility or molecular plasticity, has been hypothesized as=
 a mechanism to facilitate tight and specific interactions. To test this hy=
pothesis and to incorporate it into molecular design, we extended charge op=
timization theory to treat flexible molecules. Specifically, we framed and =
solved a single-objective optimization problem for binding affinity and a c=
orresponding multi-objective optimization problem for binding affinity and =
specificity by adjusting the charge distribution of one binding partner to =
be complementary for the other while permitting varying levels of conformat=
ional flexibility. Application to a ligand that was developed to inhibit HI=
V-1 protease identified ways to improve its binding affinity, and indicated=
 that a more favorable Pareto frontier of the binding affinity=96specificit=
y trade-off can be achieved with a more flexible ligand.<br>

<br>The second part aims to understand molecular mechanisms by which small =
molecules can exhibit binding promiscuity and to develop design strategies =
to implement such promiscuity. We chose to study the inhibition of HIV-1 pr=
otease, which remains a tremendous challenge in the face of an evolving vir=
al population. Using computational design we constructed small-molecule inh=
ibitors targeting a set of wild-type and drug-resistant mutant HIV-1 protea=
ses. Each design was solved as a combinatorial optimization problem in a di=
screte chemical and conformational space. Subsequent statistical analysis r=
evealed significant trends for promiscuous inhibitors: they tended to be sm=
aller, more flexible, and more hydrophobic compared to highly selective one=
s. Furthermore, structural analysis indicated that flexible inhibitors ofte=
n achieved promiscuity by conformational adaptations to mutations in protea=
ses. Our flexibility measure also showed its potential as a design criterio=
n for promiscuous inhibitors because inhibitors with higher flexibility mea=
sures were more likely to be promiscuous. Finally, as no inhibitor covered =
all variants, we designed small cocktails of inhibitors to do so through so=
lution of a set cover problem.<br>

<br>This talk examines two perspectives on molecular complementarity throug=
h analysis and design in an optimization framework.<br><br>-- <br>Liv Leade=
r<br>Faculty Services<br><br>Toyota Technological Institute <br>6045 S Kenw=
ood Ave, #504<br>



Chicago, IL 60637<br>Phone- (773) 834-2567<br>Fax-=A0 =A0=A0 (773) 834-9881=
<br>Email-=A0 <a href=3D"mailto:jam at ttic.edu" target=3D"_blank">lleader at tti=
c.edu</a><br>Web-=A0=A0 <a href=3D"http://www.ttic.edu/" target=3D"_blank">=
www.ttic.edu</a><br>



--002215048b973981b2049dab66a7--