Wednesday, 31 August 2011
The role of non-proprietary names (INNs, USANs etc.) and the maintenance of a synonym list for drugs under development is central in the accurate retrieval of information from the web. The USAN and INN process is well managed and documented, but it is possible to look a little bit further into the future with some simple web searches, as these examples show.
Firstly USANs under consideration - these are posted here. The therapeutic application of these can often be determined from the grammar of the name. So what compounds are they? Well a simple way to potentially look for these is with a simple whois lookup on the internet domain for INN_under_consideration.com. So using this simple approach one can speculate (reasonably) that golvatanib is Eisai's VEGFR2 and cMet inhibitor E-7050, currently in phase 2 trials. Quite a few non-proprietary domains are snaffled up by various domain registry companies, or held anonymously for their true owners - but the approach works for quite a few.
Secondly, Tradenames - well, the reference trademark site for the US is the United States Patents and Trademarks Office (USPTO). There are some excellent search tools, and one way of using these in the context of drug names is to search for the developer name, and then generating a list of potential candidates, again the field of use allows some association to be often made between a new drug. For example, searching with Pfizer as the "Owner" gives a list of trademarks including Xalkori (the tradename for the newly launched anti-cancer drug Crizotinib), and a number of alternative homonyms (these could be defensively filed, or for different territories). Of course, making a definitive assignment is difficult, and things can change, but these names could become quite useful when chained and mosaiced with other synonyms.
So, there's some simple things, it gets more interesting when you start to look at dates of filing, and so forth; but that is for another day, and a lot more analysis.
Disclaimer - the links above are fragile, they will change over time and may well have session ids, that won't work for you!
Minimum Information Standards are an important feature in many aspects of science, and there is a rich history of the success of these in encouraging data interoperability across scientific resources and data analysis. An opinion paper has just been published in NRDD, that discusses bioactivity data. The paper itself seems to be open access (from my hotel room at least) - the link is here.
%T Minimum information about a bioactive entity (MIABE)
%J Nature Reviews Drug Discovery %V 10 %P 661-669 %D 2011 %A S. Orchard %A B. Al-Lazikani %A S. Bryant %A D. Clark %A E. Calder %A I. Dix %A O. Engkvist %A M. Forster %A A. Gaulton %A M. Gilson %A R. Glen %A M. Grigorov %A K. Hammond-Kosack %A L. Harland %A A. Hopkins %A C. Larminie %A N. Lynch %A R. K. Mann %A P. Murray-Rust %A E. Lo Piparo %A C. Southan %A C. Steinbeck %A D. Wishart %A H. Hermjakob %A J. Overington %A J. Thornton %O doi:10.1038/nrd3503
Tuesday, 30 August 2011
B2R is a Rhodopsin-like receptor, 391 amino acid long, which belongs to the G protein-coupled receptor (GPCR) A3 family and is encoded by the BDKRB2 gene in humans. The amino acid sequence of B2R is:
>B2R MFSPWKISMFLSVREDSVPTTASFSADMLNVTLQGPTLNGTFAQSKCPQVEWLGWLNTIQ PPFLWVLFVLATLENIFVLSVFCLHKSSCTVAEIYLGNLAAADLILACGLPFWAITISNN FDWLFGETLCRVVNAIISMNLYSSICFLMLVSIDRYLALVKTMSMGRMRGVRWAKLYSLV IWGCTLLLSSPMLVFRTMKEYSDEGHNVTACVISYPSLIWEVFTNMLLNVVGFLLPLSVI TFCTMQIMQVLRNNEMQKFKEIQTERRATVLVLVVLLLFIICWLPFQISTFLDTLHRLGI LSSCQDERIIDVITQIASFMAYSNSCLNPLVYVIVGKRFRKKSWEVYQGVCQKGGCRSEP IQMENSMGTLRTSISVERQIHKLQDWAGSRQ
Sunday, 28 August 2011
Non-small cell lung carcinomas (NSCLC) are cancers of the epithelial cells in the lung and describe all types of lung carcinomas other than small cell carcinomas. NSCLCs make up 88% of all lung carcinomas (see Cancer Research UK pages) and comprise genetically distinct classes of cancer, the most common are: Lung adenocarcinoma, large cell carcinoma and squamous cell carcinoma. Across the NSCLC types, some tumors harbour an ALK fusion protein. The EML4-ALK fusion gene has been shown to be affect the outcome of drug response and cells show resistance to EGFR inhibitors.
Crizotinib is an orally-dosed receptor tyrosine kinase inhibitor with significant activity against ALK (UniProt:Q9UM73, CHEMBL4247), HGFR (also known as c-Met) (Uniprot:P08581 CHEMBL3717) and RON (UniProt:Q04912, CHEMBL2689), and it is the first in class agent against any of these targets. However, ALK is the main targeted protein in these ALK-dependent, EGFR inhibitor resistant cancers. ALK is a trans-membrane receptor tyrosine kinase with the regulatory domains being extracellular and the kinase catalytic domain intracellular. The amino-acid sequence of the full wild type kinase is:
>ALK (full length wild type, kinase domain in red) MGAIGLLWLLPLLLSTAAVGSGMGTGQRAGSPAAGPPLQPREPLSYSRLQRKSLAVDFVV PSLFRVYARDLLLPPSSSELKAGRPEARGSLALDCAPLLRLLGPAPGVSWTAGSPAPAEA RTLSRVLKGGSVRKLRRAKQLVLELGEEAILEGCVGPPGEAAVGLLQFNLSELFSWWIRQ GEGRLRIRLMPEKKASEVGREGRLSAAIRASQPRLLFQIFGTGHSSLESPTNMPSPSPDY FTWNLTWIMKDSFPFLSHRSRYGLECSFDFPCELEYSPPLHDLRNQSWSWRRIPSEEASQ MDLLDGPGAERSKEMPRGSFLLLNTSADSKHTILSPWMRSSSEHCTLAVSVHRHLQPSGR YIAQLLPHNEAAREILLMPTPGKHGWTVLQGRIGRPDNPFRVALEYISSGNRSLSAVDFF ALKNCSEGTSPGSKMALQSSFTCWNGTVLQLGQACDFHQDCAQGEDESQMCRKLPVGFYC NFEDGFCGWTQGTLSPHTPQWQVRTLKDARFQDHQDHALLLSTTDVPASESATVTSATFP APIKSSPCELRMSWLIRGVLRGNVSLVLVENKTGKEQGRMVWHVAAYEGLSLWQWMVLPL LDVSDRFWLQMVAWWGQGSRAIVAFDNISISLDCYLTISGEDKILQNTAPKSRNLFERNP NKELKPGENSPRQTPIFDPTVHWLFTTCGASGPHGPTQAQCNNAYQNSNLSVEVGSEGPL KGIQIWKVPATDTYSISGYGAAGGKGGKNTMMRSHGVSVLGIFNLEKDDMLYILVGQQGE DACPSTNQLIQKVCIGENNVIEEEIRVNRSVHEWAGGGGGGGGATYVFKMKDGVPVPLII AAGGGGRAYGAKTDTFHPERLENNSSVLGLNGNSGAAGGGGGWNDNTSLLWAGKSLQEGA TGGHSCPQAMKKWGWETRGGFGGGGGGCSSGGGGGGYIGGNAASNNDPEMDGEDGVSFIS PLGILYTPALKVMEGHGEVNIKHYLNCSHCEVDECHMDPESHKVICFCDHGTVLAEDGVS CIVSPTPEPHLPLSLILSVVTSALVAALVLAFSGIMIVYRRKHQELQAMQMELQSPEYKL SKLRTSTIMTDYNPNYCFAGKTSSISDLKEVPRKNITLIRGLGHGAFGEVYEGQVSGMPN DPSPLQVAVKTLPEVCSEQDELDFLMEALIISKFNHQNIVRCIGVSLQSLPRFILLELMA GGDLKSFLRETRPRPSQPSSLAMLDLLHVARDIACGCQYLEENHFIHRDIAARNCLLTCP GPGRVAKIGDFGMARDIYRASYYRKGGCAMLPVKWMPPEAFMEGIFTSKTDTWSFGVLLW EIFSLGYMPYPSKSNQEVLEFVTSGGRMDPPKNCPGPVYRIMTQCWQHQPEDRPNFAIIL ERIEYCTQDPDVINTALPIEYGPLVEEEEKVPVRPKDPEGVPPLLVSQQAKREEERSPAA PPPLPTTSSGKAAKKPTAAEISVRVPRGPAVEGGHVNMAFSQSNPPSELHKVHGSRNKPT SLWNPTYGSWFTEKPTKKNNPIAKKEPHDRGNLGLEGSCTVPPNVATGRLPGASLLLEPS SLTANMKEVPLFRLRHFPCGNVNYGYQQQGLPLEAATAPGAGHYEDTILKSKNSMNQPGPThe ELM4-ALK translocation protduces a chimeric protein with the N-terminal part of ELM4 and the catalytic region of ALK. This chimeria is constitutively active causing unregulated proliferation (Soda et al)
The structure of ALK in complex with crizotinib has been determined (2XP2)
ALK and other proteins inhibited by crizotinib are members of the large protein kinase family, the target of several other recently approved drugs - with the approval of crizotinib there are now 12 US approved small molecule protein kinase inhibitors (Imatinib, Gefitinib, Erlotinib, Sorafenib, Dasatinib, Sunitinib, Nilotinib, Lapatinib, Pazopanib, Vandetanib & Vemurafenib), with over an additional 300 protein kinase inhibitors that have progressed to clinical trials.
The molecular formula for crizotinib is C21H22Cl2FN5O, with a molecular weight of 450.34 Da. (IUPAC: (R)-3-[1-(2,6- Dichloro-3-fluorophenyl)ethoxy]-5-[1-(piperidin-4-yl)-1H-pyrazol-4-yl]pyridin-2-amine, Canonical SMILES:CC(C1=C(C=CC(=C1Cl)F)Cl)OC2=C(N=CC(=C2)C3=CN(N=C3)C4CCNCC4)N, InChI key:KTEIFNKAUNYNJU-GFCCVEGCSA-N, Chemspider:9801307, PubChem:11626560). It is fully rule of five compliant. Crizotinib has two ionizable centres with a pKa of 9.4 (for the piperidinium cation) and 5.6 (for the pyridinium cation). The experimental logD (octanol/water) at pH 7.4 is 1.65. Crizotinib is a chirally pure, synthetic small molecule drug.
Standard dosing of Crizotinib is 250mg twice daily (so a 500 mg daily dose - equivalent to 1,110 umol). The mean absolute bioavailability of Crizotinib is 43% following the administration of a single 250 mg oral dose. Crizotinib shows a median time to achieve peak concentration (Tmax) of 4 to 6 hours. Following crizotinib 250 mg twice daily dosing, steady state drug levels are reached within 15 days. The volume of distribution (Vss) of crizotinib is 1,772 L (following intravenous administration of a 50 mg dose) - so crizotinib is extensively distribution into tissues from the plasma; plasma protein binding (ppb) is 91%, and it is a substrate for P-glycoprotein (P-gp).
Crizotinib is predominantly metabolized by CYP3A4 and CYP3A5. The primary metabolic pathways in humans are oxidation of the piperidine ring to crizotinib lactam and O-dealkylation, followed by subsequent Phase 2 conjugation of O-dealkylated metabolites. Crizotinib is an inhibitor of CYP3A. The mean apparent plasma terminal half-life of crizotinib iss 42 hours, with an apparent clearance (CL/F) of 100 L/hr following a single dose, or 60 L/hr for the 250 mg twice daily standard dosing - this is most likely due to inhibition of CYP3A4/5.
Crizotinib is marketed by Pfizer, full prescribing information can be found here.
Saturday, 27 August 2011
We try to keep these current, so there will be inconsistency between the views through the ChEMBL interface and on this page, but this is currently inevitable in our versioned release of ChEMBL.
Any feedback on the view, data, etc. would be gratefully received.
Friday, 26 August 2011
One of the ideas we are considering is to have a 'pre-release' of the forthcoming release to interested parties during our normal one or two week long testing period. This could allow a smoother release schedule, and also help us with support issues after releases, but I'm sure have some further problems for us. So, if you run a resource, using ChEMBL and would like to discuss this with us, or if you have other views, we'd be happy to hear (mail link).
- Design of biopharmaceuticals using rule-based approaches.
- Drug design strategies to improve drug safety.
- Multi-scale indexing, visualisation and navigation of chemical structures.
- Drug attrition analysed using systems biology and simulation approaches.
Monday, 22 August 2011
On the August 17th 2011, the FDA approved Vemurafenib (trade name:Zelboraf TM Research code: PLX-4032, RG-7204 and RO-5185426), a BRAF kinase inhibitor for the treatment of patients with unresectable or metastatic melanoma carrying the mutant BRAFV600E.
Melanoma is a malignant tumor of melanocytes (skin cells that produce melanin) and is an aggressive disease responsible for an estimated 50,000 deaths worldwide. Over 50% of patients with advanced melanoma carry an activating mutation in the Serine/Theronine protein kinase: BRAF (V600E).
The MAPK signal transduction pathway is an important and frequently mutated pathway in cancer. A wide variety of growth factors signal through this pathway, via RAS and RAF proteins to cause cell proliferation. The activating mutation in BRAF causes activation of this pathway downstream of BRAF regardless of the presence of growth factor (the signalling pathway is 'dysregulated'). The protein of Vemurafenib is this mutant enzyme BRAFV600E (Uniprot:P15056(wt)), although it shows activity in vitro against other protein kinases including such as CRAF, ARAF, wild-type BRAF, SRMS, ACK1, MAP4K5 and FGR at similar affinities.
One side effect observed in nearly a quarter of patients is the paradoxical growth of cutaneous squamous cell carcinomas (cuSCC), a different, and less aggressive type of skin cancer. Intriguingly, this appears to be cause by activating the same pathway in normal cells of the same patient that carry a RAS mutation.
>BRAF MAALSGGGGGGAEPGQALFNGDMEPEAGAGAGAAASSAADPAIPEEVWNIKQMIKLTQEHIEALLDK FGGEHNPPSIYLEAYEEYTSKLDALQQREQQLLESLGNGTDFSVSSSASMDTVTSSSSSSLSVLPSS LSVFQNPTDVARSNPKSPQKPIVRVFLPNKQRTVVPARCGVTVRDSLKKALMMRGLIPECCAVYRIQ DGEKKPIGWDTDISWLTGEELHVEVLENVPLTTHNFVRKTFFTLAFCDFCRKLLFQGFRCQTCGYKF HQRCSTEVPLMCVNYDQLDLLFVSKFFEHHPIPQEEASLAETALTSGSSPSAPASDSIGPQILTSPS PSKSIPIPQPFRPADEDHRNQFGQRDRSSSAPNVHINTIEPVNIDDLIRDQGFRGDGGSTTGLSATP PASLPGSLTNVKALQKSPGPQRERKSSSSSEDRNRMKTLGRRDSSDDWEIPDGQITVGQRIGSGSFG TVYKGKWHGDVAVKMLNVTAPTPQQLQAFKNEVGVLRKTRHVNILLFMGYSTKPQLAIVTQWCEGSS LYHHLHIIETKFEMIKLIDIARQTAQGMDYLHAKSIIHRDLKSNNIFLHEDLTVKIGDFGLAT[V/E] KSRWSGSHQFEQLSGSILWMAPEVIRMQDKNPYSFQSDVYAFGIVLYELMTGQLPYSNINNRDQIIF MVGRGYLSPDLSKVRSNCPKAMKRLMAECLKKKRDERPLFPQILASIELLARSLPKIHRSASEPSLN RAGFQTEDFSLYACASPKTPIQAGGYGAFPVH
The structure of Vemurafenib complexed to BRAF is known (PDBe:3og7).
CHEMBL1229517, Chemspider:24747352, PubChem:CID 42611257). Vemurafenib is a synthetic small molecule drug, with no chiral centres, it has a molecular weight of 489.9 MWt and is fully rule of Five compliant.
Vemurafenib is orally administered as tablets, each tablet contains 240 mg of active compound - dosing is 960 mg twice daily (equivalent to 3920 umol). The bioavailability of vemurafenib has not been determined. Following oral administration of vemurafenib at 960 mg twice daily for 15 days to patients with metastatic melanoma, the median Tmax was approximately 3 hours. Vemurafenib is a moderate CYP1A2 inhibitor, a weak CYP2D6 inhibitor and is a CYP3A4 inducer, it is highly bound to serum albumin and alpha-1 acid glycoprotein (> 99% ppb). In treated patient populations the apparent volume of distribution is 106 L, the clearance is 31 L/day and the median half life is 57 hours. It is largely excreted in feces (94% of dose).
Vemurafenib is also notable in being arguably the first drug discovered and optimised using fragment soaking methods for initial lead discovery. Vemurafenib was discovered in the labs of Plexxikon.
Full US Prescribing information is here
Zelboraf is marketed by Hoffmann-La Roche Inc.
Sunday, 21 August 2011
Brentuximab vedotin's molecular (for the antibody component of the ADC, Brentuximab) target is CD30 (Uniprot: P28908; canSAR ; PFAM: PF00020), which is expressed on activated T- and B-cells, and also is an established tumour marker. Endogenous CD30 ligands include TRAF1, TRAF2, TRAF3 and TRAF5. CD30 is a transmembrane protein, 577 amino acid long, containing three related copies of the TNFR_c6 domain (PFAM:PF00020). The structure of TRAF2 complexed to part of CD30 is known (see PDBe:1d01)
The antibody component Brentuximab binds to CD30 expressing cells, the the complex is then internalised, and the cytotoxic agent Monomethyl auristatin E (MMAE) which blocks cell division by preventing tubulin polymerisation. Auristatin is a marine natural product peptide derivative. There are approximately 4 copies of MMAE coupled to each antibody, and MMAE is linked through a protease activatable linker. The total molecular weight of Brentuximab vedotin is ca. 153 kDa. Brentuximab can therefore be considered to have two targets (CD30 and tubulin), and also to be a prodrug.
Brentuxumab vedotin is the second US approved antibody drug conjugate - the first being gemtuzumab ozagamicin (Mylotarg) which was approved in 2001, but subsequently withdrawn from the market. A significant number of ADCs are currently in clinical development.
Brentuximab is a human antibody. The antibody portion of Brentuximab vedotin has the sequence of two copies of:
>Brentuximab vedotin - heavy chain QIQLQQSGPEVVKPGASVKISCKASGYTFTDYYITWVKQKPGQGLEWIGWIYPGSGNTKY NEKFKGKATLTVDTSSSTAFMQLSSLTSEDTAVYFCANYGNYWFAYWGQGTQVTVSAAST KGPSVFPLAPSSKSTSGGTAALGCLVKDYFPEPVTVSWNSGALTSGVHTFPAVLQSSGLY SLSSVVTVPSSSLGTQTYICNVNHKPSNTKVDKKVEPKSCDKTHTCPPCPAPELLGGPSV FLFPPKPKDTLMISRTPEVTCVVVDVSHEDPEVKFNWYVDGVEVHNAKTKPREEQYNSTY RVVSVLTVLHQDWLNGKEYKCKVSNKALPAPIEKTISKAKGQPREPQVYTLPPSRDELTK NQVSLTCLVKGFYPSDIAVEWESNGQPENNYKTTPPVLDSDGSFFLYSKLTVDKSRWQQG NVFSCSVMHEALHNHYTQKSLSLSPG >Brentuximab vedotin - light chain DIVLTQSPASLAVSLGQRATISCKASQSVDFDGDSYMNWYQQKPGQPPKVLIYAASNLES GIPARFSGSGSGTDFTLNIHPVEEEDAATYYCQQSNEDPWTFGGGTKLEIKRTVAAPSVF IFPPSDEQLKSGTASVVCLLNNFYPREAKVQWKVDNALQSGNSQESVTEQDSKDSTYSLS STLTLSKADYEKHKVYACEVTHQGLSSPVTKSFNRGEC
The linker-MMAE has the following structure:
Brentuximab is administered intravenously, and the recommended dose is 1.8 mg/kg administered over 30 minutes every 3 weeks for a maximum of sixteen cycles doses. The terminal half-life (t1/2) is 4 to 6 days.
Thursday, 11 August 2011
|Drug Type||Drug Class||Target|
|glycerol phenylbutyrate||HPN-100||synthetic small molecule prodrug||therapeutic||n/a|
|orteronel||TAK-700||synthetic small molecule drug||therapeutic||steroid 17-alpha hydroxylase/17,20 lyase|
|plazomicin||ACHN-490||natural product-derived drug||therapeutic||30S ribosome|
|radavirsen||AVI-7100||oligonucleotide derived||therapeutic||Influenza virus|
|rasagiline||TV-1030||synthetic small molecule||therapeutic||MAO-B|
|setrobuvir||ANA-598||synthetic small molecule||therapeutic||HCV polymerase|
|sutezolid||PNU-100480, PF-02341272||synthetic small molecule||therapeutic||50S ribosome|
|tozadenant||SYN-115, RO-4494351||synthetic small molecule||therapeutic||A2a receptor|
|trelagliptin||SYR-472||synthetic small molecule||therapeutic||DPP-4|
|umeclidinium, umeclidinium bromide||GSK-573719,||synthetic small molecule||therapeutic||muscarinic receptors|
Wednesday, 10 August 2011
Just a quick (but surprisingly wordy) follow up on the previous post on the drug profiling against malaria. Drug repurposing/reuse/rescue offers great potential for the enhancement of patient lives and also is a quick way of pushing new therapies through the clinic. It is often see as low cost and low risk, is highly translational in terms of the research, and there are some stunning success stories. It is therefore very sensible to screen known drugs in assays of interest, which is exactly what was done in the recent, excellent, Science paper.
The compounds in Table 1 of the paper are reported at the highly active set (and these exclude already known established antimalarial drugs which all pass the selection criteria used for compounds in this table, this seems a pretty good and pragmatic place to set cutoffs). For use as an widely-used and developing world-applicable antimalarial (co)-therapy I would have imagined that ideally you would want established well tolerated daily dosing, since existing malaria therapies are oral and have (generally) good tolerability.
So, for this list of 32 drugs, a quick internet-based classification was done - factors analysed included:
- whether the drugs were already approved for human use
- the dosage/absorption route
- any special monitoring requirements (there is probably no option for dose titration/liver monitoring for the vast majority of malaria victims).
Of the 32, I could find (using a flaky internet connection from the my holiday sun-chair) data that looked OK on 30 (the ones I couldn't easily find things on were suberoylanilide and Alazanine triclofenate - if others know anything, post something in the comments). Of these 30, five (20%) are currently animal use only, nine (30%) are IV use only, one (3%) is not currently approved in humans or animals (Lestautinib), and eight (27%) are topical/inhaled use only. This leaves six (20%) that have the desired profile of being orally dosed and currently approved for human usage. Of these, four have restrictive use/or appear to be poorly tolerated (for example, fumagilin is hospital use only and requires careful monitoring). So this leaves two reasonable candidates for first pass consideration as real 'drug-repurposing' candidates - Dextroamphetamine saccharate and orlistat.
Dextroamphetamine will clearly have regulatory and huge misuse potential, so is also probably a non-starter, regardless of any PK/exposure concerns; so Orlistat then? Orlistat is used as an over-the-counter (in some territories) and prescription drug to assist with weight loss. If works by reducing the absorption of fat from the diet, leading to a variety of side effects - there is much discussion on the safety and side effects (google will find a lot of comment), so in populations that are in malaria endemic regions, a drug that could be dosed to underweight malnourished patients and that reduce the absorption of a key food group (and also most fatty essential vitamins) may not be optimal. A larger issue though is that orlistat works in the lumen of the stomach and intestine, where the target human enzyme is secreted, so for obesity treatment, orlistat does not need to be absorbed into the rest of the body, and it isn't. So, ignoring the side effects for a moment, the negligible systemic exposure of orlistat again probably precludes its use as an antimalarial.
It will be interesting to me if this sort of pattern of results, and the importance of considering in vitro to in vivo translation for 'known drug' screening is typical. Sorry for the long post - thoughts, comments and discussion most welcome.
Update: Looking at the supplementary material for the paper - it seems likely that suberoylanilide from table 1 is actually suberoylanilide hydroxamic acid (aka Vorinostat, or Zolinza). This is an oral drug for the treatment of certain refractory cancers, it can give rise to a wide range of serious side effects that would almost certainly preclude widespread non-supervised use (at least at the doses used for current indications).
Update 2: Thanks to the authors for confirming that the suberoylanilide should be SAHA, and also for confirming that Alazanide triclofenate is an unusual compound with not a lot of current published literature.
- 1,195,368 compound records
- 1,060,258 distinct compounds
- 582,982 assays
- 5,479,146 activities
- 8,603 targets
- 42,516 documents
- 7 activity data sources
- The loading of the quantitative activity data from the Guide to Receptors and Channels (4th Edition)
- Updating of the organism classification information
- The ChEMBL identifiers for all previous compounds/assays/documents/targets within the database, including those that have been removed/downgraded are now maintained within the chembl_id_lookup table
- An InChI key resolving URL has been set up, e.g. https://www.ebi.ac.uk/chembl/compound/inchikey/NXQMNKUGGYNLBY-UHFFFAOYSA-N
- You can now search for a list of target ChEMBL identifiers on the target search page, https://www.ebi.ac.uk/chembl/target