Biomedical Engineering Reference
In-Depth Information
object-oriented software development, strong Web standards adherence, thor-
ough automated software testing, and horizontal scaling via a shared-nothing
architecture. CDD's unusually disciplined approach to automated software
testing and a well-factored code base translate into a nearly bug-free user
experience for its growing user community which interacts with millions of
compounds and tens of millions of data points on a daily basis.
The CDD database can archive and mine a broad range of diverse objects
that can later be selectively and securely shared with other researchers (or
permanently kept private, which is the default behavior). The CDD database
is a hosted collaborative system with an important advantage over traditional
PC-based database systems since it can enable secure login into the database
from any computer using any common browser (e.g., Firefox, Internet Explorer,
Chrome, or Safari). This unique capability for a database system provides fl ex-
ibility for the users. The CDD Web-based database architecture handles a
broad array of data types and is arranged as three specifi c modules. First, the
CDD Vault securely stores and enables mining of private data which are
hosted and managed by CDD. Second, CDD Collaborate enables confi dential
exchange of data between vaults as selected by users. Third, CDD Public hosts
public data sets that can be mined. The CDD platform incorporates Marvin,
calculated plug-ins for physical chemical calculations, and the JChem Cartridge
for structure searching from ChemAxon (Budapest, Hungary) within the
application as the chemistry engine. This allows one to do sophisticated
structure-activity relationship (SAR) analysis, including chemical pattern rec-
ognition (e.g., similarity and substructure searching), physical chemical prop-
erty calculations, Boolean search and save capabilities for potency, selectivity,
toxicity, and other experimentally derived properties. The database can handle
heterogeneous data fi les as well as standardized csv and sdf fi le convertible
formats that represent the chemical and biological data. In particular, CDD is
tailored for common data formats used by biologists such as Microsoft Excel
(.xls) and text (.txt) fi les. The technology can mine against a variety of values,
including concentration, time, percent, real, integer, textline, cpm, rlu,
Z
/
Z
′
plate statistics, and IC
50
(log IC
50
,
R
2
values, Hillslope, etc.). The outputs of such
mining can be saved, exported, shared inside CDD, or plotted with an inte-
grated plug - in.
The researcher can control which data to keep 100% private, share with
groups of individual researchers, or share more generally with the public. A
further unique capability of CDD is the ability to compare all or subsets of
public access data with private data simultaneously in a single container as
well as analyze multiple vaults to which the user has access. The power of
this collaborative approach to drug discovery can be seen in different types
of community-based research projects. These range from traditional com-
pletely private collaborations, to temporally private collaborations which may
become more open following a privacy escrow period, to completely open
collaborations where researchers can blog about the experiments as they
occur [19] .
Search WWH ::
Custom Search