Bioprinter

Figure 2.1 BioPrint approach the in vitro data are all measured in a consistent manner with full dose response for any activity >30% at 10pM the in vivo data are curated from available data, supplemented by custom measured data from collaborators.

Figure 2.2 BioPrint assays (cyan) mapped onto the drugable proteome.

In this chapter, the full BioPrint approach is described, as available from Cerep in terms of both the data set and the ability to have new compounds profiled and the results provided in the context of the BioPrint data set, including the known in vivo side effects of near neighbors in this biological space (see Section 2.5). The results for the differentiation of hit/lead compounds (see Section 2.3.2.1) sometimes use a subset of the 70-100 pharmacological assays that provide the maximum signal. Usually a decision on future work prioritization could be clearly made from the data from these subsets, saving time and money. For key reference/tool compounds, a full profile was used and is recommended to be used, as unexpected off-target activities may be found that cannot usually be predicted. 

BioPrint ] database, plC50]. Biological assays are ofthe activities of the same set ofcompounds for on the x-axis, and compounds on the y-axis each target. The activities are shown in the form (i.e. a row contains the biological fingerprint of a of a heat map, with red most active and blue-compound as heat map). Hierarchical clustering green inactive. Compounds tend to cluster into has been performed on both axes compounds therapeutic area and this is marked on the left. 

Figure 2.6 shows a histogram of a number of different targets (assays) against which drugs (1388) in the BioPrint database were found to be active (using a cutoff of 50% inhibition at 10 pM). 

Figure 2.6 Compound target activity distribution ( promiscuity ) for 1388 drugs profiled in BioPrint assay panel (with 50% inhibition at 10pM taken as active), shown as a histogram. The number of active compounds is shown along the y-axis and the number of targets along the x-axis (adapted from ref. [6]).

Figure2.7 Selectivity ( promiscuity ,x-axis) intermsofcompound target activity numbers for 1098 drugs from BioPrint profiled in the BioPrint assay panel (with <1 pM IC50 taken as active), versus clogP (hydrophobicity, y-axis) (adapted from ref. [5]).

BioPrint consists of a large database and a set of tools with which both the data and the models generated from the data can be accessed. The database contains structural information, in vivo and in vitro data on most of the marketed pharmaceuticals and a variety of other reference compounds. The in vitro data generated consist of panels of pharmacology and early ADME assays. The in vivo data consist of ADR data extracted from drug labels, mechanisms of action, associated therapeutic areas, pharmacokinetic (PK) data and route of administration data. 

The BioPrint profile has been set up to deliver early assessment of the potential off-target and ADME(T) properties of lead compounds. To quickly provide coherent data... 

Collaborators have profiled their own compounds in the identical assays, to produce enhanced proprietary data sets. These are generally project and attrited compounds. Examples of results using data from these compounds in the context of BioPrint compounds are given in Section 2.3.2.1 for the differentiation ofhits/leads and in Section 2.3.2.2 for the analysis of attrited compounds. 

BioPrint also includes a small animal toxicity data set. Blood chemistry and organ toxicity data have been gathered for more than 200 compounds. 

To demonstrate this, Figure 2.8 shows the comparison of similarity for Daylight structural and biological fingerprints created from a panel of 154 assays from the BioPrint database (measured by pairwise Tanimoto distance for 347 drugs with MW 200-600 60031 points) [6]. Figure 2.8a shows the overall scatter plot of the... 

As illustrated in the next section, the use of biological fingerprints, such as from a BioPrint profile, provides a way to characterize, differentiate and cluster compounds that is more relevant in terms ofthe biological activity of the compounds. The data also show that different in silico descriptors based on the chemical structure can produce quite different results. Thus, the selection of the in silico descriptor to be used, which can range from structural fragments (e.g. MACCS keys), through structural motifs (Daylight keys) to pharmacophore/shape keys (based on both the 2D structure via connectivity and from actual 3D conformations), is very important and some form of validation for the problem at hand should be performed. 

An early identification of the best leads is critical, and systematic biological profiling, as with Cerep BioPrint , enables this progress to a great extent. Experience with using this approach in a major pharmaceutical company (Pfizer) has confirmed this. One component of the issue can be stated that it is better to identify the best (not just potency) lead series than to try and find the best candidate from a possibly suboptimal series, which can happen at late stages of lead optimization, where it is very difficult to make major changes to the lead series chemistry. 

Figure 2.10 Clustering of hit/lead compounds based on biological fingerprints along with a clinical reference compound Duloxetine in the context of the BioPrint data set. The rows show the partial heat map of activity for a compound (assays on the x-axis). Activity is color coded from red (very active) to blue-green (inactive).

Figure 2.12 (a) Analysis of >130 attrited compounds using the BioPrint assays. The 15 targets hit by more than two attrited compounds (>50% inhibition at 10 pM) are shown, (b) Activity of the drug compound set against the 15 assays shown in (a). 

Figure 2.13 BioPrint (partial) profile of compounds developed for activity against a serotonin receptor. Three compounds that attrited for some type of toxicity issue are shown at the bottom in a purple box. The serotonin receptors and transporters are highlighted with a blue box.

Figure 2.14 BioPrint (partial) profile of some HMGCoA inhibitors. The two main structural classes are highlighted by a blue and purple box for both the activity fingerprint and the 2D structures.

Figure 2.15 BioPrint (partial) data for a set of compounds developed against the same enzyme target.

---