Table 1.
Content statistics of MCD and TCD, populated from journal papers and patents, respectively.
Figure 1.
Compounds linked to human protein targets.
These were extracted by year from patents (red) and journals (blue).
Figure 2.
Patent office document counts.
The light blue and orange plots are data from WIPO IP Statistics Data Centre [50]. These were selected as total world-wide patent applications for all categories (orange WO all, but divided by 10 for scaling) or the technology category pharmaceuticals (blue WO pharma). The other two plots were generated using the Espacenet advanced search interface [51]. With the selection of the IPC code C07D the red line is the total and the green line added a selection for “university OR institute” in the applicant field. For 2010 the numbers retrieved were 47,884 for C07D, 1,189 of these with “institute” and 2,366 with “university”.
Figure 3.
Average compounds-per-document by year from patents (red) and journals (blue).
Figure 4.
These are the additive totals up to each year, for proteins specified in patents (red) and journals (blue) from all species.
Figure 5.
Each total represents the target proteins specified in the patents (red) and journals (blue) published in that year.
Figure 6.
Compounds published by GlaxoSmithKline.
Counts by year in patents (red) and journals (blue).
Figure 7.
GlaxoSmithKline human targets counts.
As extracted from patents (red) and journals (blue) published by year.
Table 2.
Ranking of the top-ten targets by year.
Figure 8.
Time courses for DPPIV (UniProt P27487).
(A) Shows the number of compounds (vertical axis) linked to individual proteins from patents (red) and journal articles (blue). (B) Is the average MW for compounds (vertical axis) each year from patents (red) and journal articles (blue). Missing points are years where no documents were extracted. (C) Shows queries in PubMed. Abstracts returned by “dpp iv inhibitors” are plotted by year (blue bars). The red bars are clinical trial reports. The arrowed time points are as follows: 1994 early dipeptide inhibitors [52], 1998 the first review related to diabetes therapy [53], 2002 the first clinical trial results [54], and in 2006 the approval of sitagliptin.
Figure 9.
Time courses for renin (REN, UniProt P00797).
The horizontal axes and line labels are the same for Figure 8 except for the vertical scales. (A) Shows the number of compounds linked to renin. (B) Shows the average MW for compounds. (C) shows queries in PubMed (in this case “renin inhibitors” needed to be constrained as a phrase query to preclude “renin-angiotensin system” false positives). The arrowed time point is the 2007 FDA approval of aliskiren.
Figure 10.
Time course for thrombin (F2, UniProt P00734).
The horizontal axes and line labels are the same for Figure 8 except for the vertical scales. (A) Shows compounds linked to thrombin. (B) shows average MW for compounds. (C) Shows queries in PubMed for inhibitors and clinical trials (but the pre-2003 and some subsequent results are associated with hirudin analogues and other non-small molecule inhibitors). The arrowed timelines are the approval of atroban as an injectable in 2000, the first clinical trial of an oral inhibitor, ximelagatran, in 2003 and the FDA approval of dabigatran in 2008.
Figure 11.
Time courses for Factor X (F10, UniProt P00742).
The horizontal axes and line labels are the same as for Figure 8 except for the vertical scales. (A) Shows compounds linked to F10. (B) shows average MW for compounds. (C) Shows queries in PubMed for inhibitors and clinical trials the results are associated with heparin analogues and other non-small molecule inhibitors. The arrowed timelines are for the first PDB structure in 1993 and first clinical trials of direct inhibitors in 2005.
Figure 12.
Time courses for Trypsin (PRSS1, UniProt P07477).
The horizontal axes and line labels are the same for Figure 8 except for the vertical scales. (A) Shows compounds linked to Trypsin. (B) shows average MW for compounds.
Figure 13.
Time courses for BACE1 (UniProt P56817).
The horizontal axes and line labels are the same for Figure 8 except for the vertical scales. (A) Shows compounds linked to BACE1. Note here the profile for papers looks low because it was plotted on the same scale as used for patents. (B) shows average MW for compounds. (C) Shows queries in PubMed for BACE1 inhibitors; those before 1999 are spurious because, at default settings, gene name queries record back-mappings, in this case investigations on inhibiting beta-secretase activity where the MeSH system added the BACE1 gene term retrospectively. The arrowed timelines are for the identification of BACE1 in 1999 and first PDB structure in 2000.
Figure 14.
Time courses for BACE2 (UniProt Q9Y5Z0).
The horizontal axes and line labels are the same for Figure 8 except for the vertical scales. (A) Shows compounds linked to BACE2. (B) shows average MW for compounds. There are no specific inhibitor publications retrieved with PubMed queries.