279Bioinformatic Approach to Personalizing Treatment of GBM
An example of bioinformatic approach to personalize treatment of GBM, is IBM’s
collaboration with New York Genome Center (NYGC) to use its Watson supercom-
puter to make sense of vast amounts of genetic sequencing data and medical infor-
mation for identifying personalized treatments for cancer patients. The partners will
test a Watson prototype designed specifi cally for genomic research to help oncolo-
gists deliver unique, customized cancer treatments according to an individual
patient’s DNA. A search of alternative treatments for glioblastoma multiforme will
test this approach.
GBM patients at the NYGC’s member institutions will be selected for the Watson
study. Each patient’s tumors will be sequenced at the Genome Center on Illumina
servers running algorithms in the IBM SoftLayer cloud. Biopsies are conducted on
patients, and both normal and cancer cells are sequenced by the NYGC’s servers.
The sequencing normally takes 10–12 days because of the intricacy of the task;
regular cells have to be sequenced ~30 times and cancer cells 30–50 times. In the
slow-but-groundbreaking process, algorithms developed through years of public
and private sector research create perfect representations of the patients’ cells in bits
and bytes.
In the next stage, which takes a few weeks, the raw sequences for healthy and
cancerous cells are extrapolated and put through heuristic algorithms to fi gure out
what healthy and cancerous cells look like in each patient. This information is used
to create variant call fi les − raw info fi les used by the NYGC’s software to store gene
sequence variations. These fi les are what Watson uses to fi nd novel cancer treat-
ments. Each variant fi le can contain between 20,000 to 1 million potential muta-
tions. Among them is a driver mutation that primarily fuels the cancer, and passenger
mutations that have much less effect. Watson combines fi ndings from the NYGC’s
programs with automated queries of a massive medical text database to attempt to
identify the driver mutation, which would be the target for personalized treatment.
Biosimulation Approach to Personalizing Treatment of Brain Cancer
Gene Network Sciences (GNS), using its REFS™ (Reverse Engineering and
Forward Simulation) technology, is collaborating of with M.D. Anderson Cancer
Center (Houston, TX) to translate DNA sequence and clinical data from GBM
patients into breakthrough discoveries leading to drugs and diagnostics. The results
from these projects will include the identifi cation of new combination drug targets
for disease and the development of diagnostics to determine appropriate individual
patient treatments. The parties plan to transform this coherent clinical 3D Data into
computer models which link genetic alterations to changes in gene expression to
progression-free patient survival times. This computer model, developed by using
the REFS™ platform, is expected to unravel the complex genetic circuitry underly-
ing GBM and reveal novel drug targets and biomarkers of response. These targets
and biomarkers may be used to identify the optimal single or combination drug
Personalized Management of Cancers of Various Organs