In2Fab (www.in2fab.com) has the only commercial
solution speciallized for general porting of analog and/or mixed signal
designs from one process generation to the next or one foundry to another,
or both at the same time. They provide either a design porting service
at their site, or they sell their productized EDA tools to enable customers
to port their own designs. This technology now has about 6 years of
high first silicon success rate history on more than 100 designs, and
is silicon and production proven on a wide range of processes at semiconductor
foundries and private captive fabs from from 0.5 micron through 65nm
processes.
What particularly standsout about their products is that they can port
typical analog or mixed signal designs such as PLLs, DACs, ADCs, etc.
in a very short 2 to 3 weeks time, and complex analog designs such as
PCI express in as little as 7 or 8 weeks. There are no complexity limits
found yet. This technology has been used to port complex SOCs including
digital in as little as 4 to 6 weeks.
The technology faithfully preserves hierarchy, and topology from source
process to target process and supports mapping of passive devices such
as resistors and capacitors to the same value on the target process
thereby preserving reference.
The technology supports Cadence style p-cells and can remap from source
to target processes and help clean up discontinuities that result. It
can also port GDSII layout data. The technology will port both the netlists
and layout. The EDA tools use a pattented technology called complex
nodal scaling, which is significantly more than just optical scaling.
The EDA tools have extensive support for complying with the increasingly
complex "design for manufacturability" and "design for
yield" rules that are prevalent in the advanced processes today.
Portland Bioscience (www.pdx-bio.com)
offers the only bioinformatic experimental design and analysis technology
that comprehensively incorporates the impact of multiplex hybridization.
This applies to all possible sample target and probe combinations. This
is achieved by a combination of theoretical modeling of molecular chemistry
and calibration with the only comprehensive thermodynamic database of
mismatch, cross hybridization taxonomies and stabilities. There is no
other company with this technology.
With each PdxBio project,
results are captured and the database is continually expanded, resulting
in the accretion of high informatics value. New types of data can be
generated and employed to expand the database to include ever more complex
molecular characteristics and chemical and physical behaviors. The technology
is composed of a set of toolbox software routines and the database that
leverages over 10 years of research and five years of PdxBio technology
development. These tools can be rapidly deployed to provide a wide range
of services and software products. The toolbox algorithms and database
searching techniques are extremely fast and, unlike existing search
and comparison technologies, primarily search for dissimilarities rather
than similarities in hybridization interactions. This greatly enhances
the quality of the results.
Impact on prior State-of-the-Art:
It is often quoted that the state of the art in microarray design and
analysis technology is between 60% and 99% accurate. In fact, the situation
is much more complicated than being able to simply state an accuracy
number. Today¡¯s technology either largely ignores complexities associated
with multiplexing such as cross-hybridization, or does so in a very
crude or limited manner. Experimental results may produce erroneous
positive calls because of an accumulation of cross-hybrids on various
probe spots. Some lab tests may be based upon only a few specific calls,
so an erroneous positive could significantly obscure results. Additionally,
a probe spot could be displaying a true positive, but may show a relatively
weaker signal due to reduction of the target concentration which results
from extensive cross-hybridization with other probes on the surface.
Some cross-hybrids can appear to be even more stable than a perfect-match
duplex. Consequently, a researcher may be unable to determine whether
or not a reliable result has been obtained. A specific test may work
well with one target sample, and be completely erroneous on a second
sample. In the interpretation of final results, current technologies
ignore the mutual influence of all possible reactions in the multiplex
hybridization system.
There are many new high-throughput
platforms that are either in development or being piloted. These will
generate extremely large amounts of data which will require the type
of system-wide analytical approaches provided by PdxBio¡¯s Multiplex
Cross-Hybridization technology.
With PdxBio technology, a
researcher can either design probes that will not produce erroneous
signals with a specific target sample, or can premodel, or post model
an experiment. This will enable the researcher for the first time to
develop a quantitative understanding of the outcome of their multiplex
hybridization reactions.