Biomedical Engineering Reference
In-Depth Information
detectable faults from Case 2 are modified with the new drug weights. In addition,
the output of the circuit is fixed to the best output as determined in Case 2. These
circuits are then solved using WPMS to obtain the selected drugs with lowest cost.
5.3.4.4
Case 4: Determining Therapy with Fewest Drugs and Best Coverage
From Case 2, we identify the drug selection that best rectifies a
certain
fault. However,
in drug therapy, the fault location may be unknown. In this situation, a drug selection
that rectifies all faults (or as many faults as possible) with the fewest drugs, is
desirable.
For each faulty circuit (with a single fault), we find all combinations of 1, 2, and
3 drugs that yield the best output from Case 2. This is done by performing a WPMS
All-SAT to find
all
satisfying drug selections with drug weight greater than or equal
to
d
3, where
d
is the total number of drugs. Each drug selection (or vector) is
analyzed to see how many testable faults are rectified or covered by it. The drug vector
with the highest coverage and fewest drugs is recorded as a best candidate for therapy.
−
5.4
Results
5.4.1
Model Implementation
We evaluate the WPMS-based ATPG methods on the GRN that models growth factor
(GF) pathways [
5
]. In multicellular organisms, cell growth and replication is tightly
controlled by the cell cycle control. This system receives signals from other cells
which are used to decide whether the cell should grow. A failure in these signals
can lead to unwanted or unregulated cell growth, leading to cancer. These signaling
pathways are well studied, and several drugs have been developed to target different
pathways for cancer therapy.
We begin with a BN model of the GF pathways as derived in [
5
]. In this model,
pathways are converted to an equivalent BN logic gate. Each interconnection (net)
between logic gates is then assigned a numerical label.
As stated in our approach section, defects in the GRN are represented as stuck-
at faults that permanently set a signal net to 1 or 0. At each net, the logic gates
for injecting a s-a-0 or s-a-1 are inserted. If there is a drug that targets the net,
the appropriate logic gates are also inserted. The conversion of the faults and drug
locations to a logic netlist is shown in Fig.
5.3
. The final circuit is then converted to
CNF for further analysis.
In the results, stuck-at faults are referred by the net numbers that are affected (i.e.
net 7 s-a-0, means that the signal corresponding to net 7 is stuck-at 0). The network
has 5 primary input (PI) signals and 7 primary output (PO) signals. The PIs will be
defined as a 5-bit binary vector:
X
=
[
EGF, HBEGF, IGF, NRG1, PTEN
]
