Biomedical Engineering Reference
In-Depth Information
(several days vs. 10 to 24 h for Roche's 454), the Illumina platform generates approx-
imately 40 times more sequencing data (about 20 Gb) per sequencing run at a similar
cost [108-111]. For reasons such as these, Illumina technology quickly became the
industry standard for DNA-encoded chemical library decoding.
In general, high-throughput-sequencing decoding simply requires the PCR inser-
tion of a platform-specific DNA adapter into the selection amplicon sequences. There-
fore, PCR bar coding of individual selection experiments is often used to maximize
the capabilities of the sequencing throughput, facilitating the parallel analysis of
multiple selections in a single sequencing run.
Depending on the library format, sequence counts obtained after deconvolution
of sequencing results can conveniently be plotted as depicted in Figure 11.22a to c.
However, statistical analyses are required to accurately describe the distribution of
library sequences before and after the selection process [56]. For example, quantile-
quantile plots and fitted negative binomial density functions allow the determination
of
p
values and of other significant statistical parameters, crucial for the reliable
identification of the compounds enriched (Figure 11.22d) [56].
As DNA-encoded chemical libraries are growing steadily in size [58], alterna-
tive high-throughput-sequencing technologies may also be considered (e.g., SOLiD
sequencing [108]) for efficient library decoding. Perhaps advances in technologies
based on single-molecule sequencing [114] may further facilitate the processing of a
very large DNA-encoded chemical library of up to 1 billion chemical entities in just
a few days.
11.4 DRUG DISCOVERY BY DNA-ENCODED CHEMICAL LIBRARIES
Over the past decade, several pharmaceutical companies and academic laboratories
have integrated DNA-encoded strategies in the de novo discovery of small molecules
capable of selective binding to specific biological targets. Table 11.1 summarizes
a number of such applications as they have been reported by some of the major
companies and academic groups currently active in this field.
Remarkable efforts in the implementation of DNA-encoded chemical library tech-
nology for drug discovery purposes have been reported by GlaxoSmithKline (for-
merly Praecis Pharmaceuticals) and by Philochem AG in collaboration with ETH
Zurich.
In 2008, Dumelin et al. at Philochem described the use of DNA-encoded chem-
ical libraries for the identification and characterization of a novel class of highly
specific portable albumin binders based on 4-(
p
-iodophenyl)butyric acid structure,
with affinities to human serum albumin ranging from 3.2 to 55
M (entry 9, Table
11.1) [64]. Conjugates of the highest-affinity ligand to a general carbonic anhydrase
inhibitor (acetazolamide) or to antibody fragments (specific for tumor-associated anti-
gens) have been demostrated to improve the pharmacokinetic properties of the small
organic molecules [96] as well as the tumor uptake of tumor-targeting antibodies
[98]. Additionally, conjugation of the newly discovered albumin binder to blood pool
contrast agents (fluorescein and Gd-DTPA) showed a significantly increase in their
