ChEMBL Resources

Resources:
ChEMBL
|
SureChEMBL
|
ChEMBL-NTD
|
ChEMBL-Malaria
|
The SARfaris: GPCR, Kinase, ADME
|
UniChem
|
DrugEBIlity
|
ECBD

Wednesday, 27 May 2009

Industry Partner Druggability Workshop

On Friday July 10th we plan to hold a further Druggability Workshop for the EMBL-EBI Industry Partners. If anyone is interested in participating in this highly interactive and participatory workshop AND also interested in joining the Industry Partner program; please contact me.....

Thursday, 14 May 2009

New Drug Approvals - Pt. III - Golimumab (Simponi)

Third on our series of posts on new FDA drug approvals this year is Golimumab, approved on the 24th of April. Golimumab is a tumor necrosis factor alpha (TNFα) blocker indicated for the treatment of active forms of rheumatoid arthritis, psoriatic arthritis and ankylosing spondylitis. Golimumab is the fourth TNF inhibitor, after Etanercept, Infliximab and Adalimumab, to reach the market. Etanercept is a fusion protein of an engineered from the the TNF receptor fused to IgG1, while Infliximab and Adalimumab are very similar to Golimumab structurally in that they are all 'conventional' monoclonal antibodies. Golimumab (previously known as CNTO-148) is a human monoclonal antibody that exhibits multiple glycoforms with molecular weights of ca. 151 kDa. Golimumab has a good subcutaneous (sc) absorption (ca. 53% bioavailable), a plasma half-life of ca. 2 weeks, a volume of distribution of 58 to 126 mL/kg and a systemic clearance of 4.9 to 6.7 mL/day/kg. The recommended dosage of 50 mg is administrated by subcutaneous injection just once a month. The full prescribing information can be found here.

Golimumab has a boxed warning (colloquially known as a 'black box').

Golimumab is a monomeric immunoglobulin (IgG) consisting of four polypeptide chains in a "Y"- shaped form: two identical heavy chains of ~450 aminoacids and two identical light chains of ~217 animoacids, connected by disulfide bonds.

Golimumab molecular formula: C6530H10068N1752O2026S44

Golimumab CAS registry: 476181-74-5

The license holder for Golimumab is Centocor Ortho Biotech Inc. and the product website is www.simponi.com.

Sunday, 10 May 2009

New Drug Approvals - Pt. II - Febuxostat (Uloric/Adenuric)

Second onto market this year is Febuxostat, approved on the 13th of February. Febuxostat is a xanthine oxidase inhibitor used for the treatment of hyperuricemia (gout), and competes against well established agents such as Allopurinol (Allopurinol was approved in 1964 for the treatment of gout). Febuxostat is a small molecule drug (Molecular Weight of 316.4 g.mol-1) is fully Rule-Of-Five compliant, is essentially insoluble in water, and has good oral absorption (>49% bioavailable). Febuxostat plasma half-life of 5 to 8 hours, a volume of distribution of 50L, high plasma protein binding at 99.2% (unsurprising given the lipophilic acid nature of the molecule). Febuxostat is extensively metabolised, both by the UGT system, producing glucoronidated metabolites, and also by a range of complex CYP mediated transformations (metabolites of metabolites....). Some of the CYP mediated metabolites involve hydroxylation of the butyl sidechain, these themselves are pharmacologically active against the target. Roughly half the dose is cleared renally, and the other half in the faeces. Typical daily dosing is 40mg (or ~126µmol), or 80mg if insufficient efficacy is observed. The full prescribing information is here.

The structure (2-[3-cyano- 4-(2-methylpropoxy) phenyl]-4-methylthiazole-5-carboxylic acid) contains a nitrile group, and a free carboxylic acid group (which will make the drug negatively charged under physiologic conditions and dominate its physical chemistry). The molecule is quite rigid and comparatively planar (due to the presence of sp2 hybridized atoms adjacent to the core phenyl and thiazole rings). A nitrile is quite unusual in drugs due to its comparative susceptibility to nucleophilic attack, and also metabolic stability - nitriles can be hydrolysed to carboxylic acids.

Febuxostat canonical SMILES: CC1=C(SC(=N1)C2=CC(=C(C=C2)OCC(C)C)C#N)C(=O)O Febuxostat InChI: InChI=1S/C16H16N2O3S/c1-9(2)8-21-13-5-4-11(6-12(13)7-17)15-18-10(3)14(22-15)16(19)20/h4-6,9H,8H2,1-3H3,(H,19,20) Febuxostat InChIKey: BQSJTQLCZDPROO-UHFFFAOYSA-N Febuxostat CAS registry: 144060-53-7 Febuxostat Chemdraw: Febuxostat.cdx

The manufacturer of Febuxostat is Takeda Pharmaceuticals and the product website is www.uloric.com

New Drug Approvals - Pt. I - Milnacipran (Ixel/Savella)

An occasional series of posts with news and top level details on new FDA drug approvals. The series is a little retrospective at the moment, but it will not take much time to catch up for 2009.

First 'out of the gates' on 14th Jan this year is an 'old' compound, Milnacipran (Savella in the US) for the treatment of fibromyalgia syndrome. Milnacipran is an SNRI (Serotonin Norepinephrine Reuptake Inhibitor, and as such blocks the function of the norepinephrine transporter NET and the serotonin transporter SERT), like all compounds of this pharmacological class, the molecule is small (Molecular Weight of 246.4 g.mol-1 for Milnacipran itself, and 282.2 g.mol-1 for the HCl salt) is fully Rule-Of-Five compliant, has high aqueous solubility, and also has good oral absorption (85% bioavailable) and metabolism characteristics. (A plasma half-life of ca. 8 hours, a volume of distribution of 400L, low plasma protein binding at 13%. Excretion is renal (through the kidneys and urine) and excretion is primarily of the unchanged drug (55% unchanged). With a molecule like this, it should come as no surprise that it is delivered orally. Recommended dosage is 100mg (or ~354µmol, or ~177µmol of the pharmacologically active enantiomer, see below) per day, although dosing is phased from a low dose to 'full' dose over a week. Milnacipran is more potent at blocking norepinephrine uptake than serotonin uptake. The full prescribing information is here.

Milnacipran has a boxed warning (colloquially known as a 'black box').

Notable features of the chemical structure ((±)-[1R(S),2S(R)]-2-(aminomethyl)-N,N-diethyl-1- phenylcyclopropanecarboxamide hydrochloride) are the primary amine (the NH2) which is basic (it can be protonated and therefore positively charged), which will dominate its physicochemical properties. The remainder of the molecule is quite lipophillic. Of further note is the cyclopropane ring (the triangle part in the middle), which is completely rigid and will dominate the three-dimensional properties (shape) of the drug. The functional groups attached to the cyclopropane are attached with mixed stereochemistry, so the drug is 'racemic' - for this reason stereochemistry is not explicitly shown on the the 2-D structure below. However, the stereoisomers have different pharmacological activity, with the 'active' d-isomer having different properties (e.g. a longer half-life) to the 'inactive' l-isomer Finally, the approved drug is not Milnacipran itself, but the Hydrochloride salt (so in the tablet the amine is indeed protonated, as discussed above).

Milnacipran canonical SMILES: CCN(CC)C(=O)C1(CC1CN)C2=CC=CC=C2 Milnacipran InChI: InChI=1S/C15H22N2O/c1-3-17(4-2)14(18)15(10-13(15)11-16)12-8-6-5-7-9-12/h5-9,13H,3-4,10-11,16H2,1-2H3 Milnacipran InChIKey: GJJFMKBJSRMPLA-UHFFFAOYSA-N Milnacipran HCl CAS registry: 101152-94-7 Milnacipran CAS registry: 92623-85-3 Milnacipran (two enantiomers) Chemdraw: Milnacipran.cdx

I mentioned it was an old compound, well it is, just that old is relative. The SNRI/SSRI therapeutic classes are old ones, and several drugs of these related classes are now available generically. SSRIs and SNRIs are primarily 'antidepressant' drugs, and this compound itself has been used in certain countries for some time (Austria, since 1998, and also at least Chile and Israel), there has been much interest in expanding the clinical use of SSRIs/SNRIs beyond their original antidepressant actions. As one would expect, the commercial history of Milnacipran appears quite complicated, and the latest approval is for a 'new' therapeutic indication - fibromyalgia (chronic and widespread muscle pain). Fibromyalgia is a complicated disease with several competing theories for its causes, and is often associated/comorbid with depression.

The license holder for Milnacipran is Cypress Bioscience and the product website is www.savella.com

Finally, given the age of the compound, it would be interesting to know what the patent status of the compound is?

Wednesday, 6 May 2009

RSC Chemical Tools Meeting, October 1st 2009, Stevenage, Herts

Here is a link for a Royal Society of Chemistry (RSC) meeting held in Stevenage in October.

The schedule is....

10.00 Herbert Waldmann MPI of Molecular Physiology, Dortmund, Germany
Bioactivity Guided Navigation of Chemical Space

10.40 John Overington European Bioinformatics Institute, Hinxton, UK
ChEMBL - Open Access Databases and Tools for Drug Discovery

11.20 Malcolm Young University of Newcastle, UK
Counterintuitive properties of complex networks in brains and cells

12.00 Lunch

13.00 Chris Schofield University of Oxford, UK
The Hypoxic Response - from Molecules to Systems

13:40 Albert Heck University of Utrecht, The Netherlands
Chemical Proteomics: Exploring the Role of Small Molecules in Molecular Systems Biology

14.40 Seth Grant MRC Cambridge, UK
Synapse Systems Biology: New Models of Brain Organisation

15.20 Glenn Prestwich University of Utah, Salt Lake City, USA
Engineered Extracellular Matrices for Drug Discovery and Reparative Medicine

16.00 Round Table Discussion

17:00 Close

The picture above is of Stevenage town center, I have never seen the sky so blue there (I live nearby), maybe it is photoshopped?

SMR Meeting: Optimising Locally Delivered Drugs

The Society for Medicines Research (SMR) is a UK-based society for those interested in drug research, they hold a regular series of meetings themed around particular areas.

When I got home yesterday, in my letterbox,I found the flyer for their next meeting, which looks good, so I thought I would post details here, and potentially encourage people to either go, or maybe even join the SMR.

The next meeting focussing on site-specific delivery of particular therapeutic agents and principles of local drug delivery is on the 11th June and is at Horsham in West Sussex. The program and registration form can be found here.

The picture above I found on the web, It is of a 'letterbox' letterbox, sort of recursive, eh?

Monday, 4 May 2009

StARlite schema walkthrough

Now that is one cool tattoo! Hey, "StARlite" has eight letters too, that gives me a good idea for a prize in a future competition....

The next StARlite schema walkthrough will be held on Friday 8th May, at 11am BST. It will last about 45 minutes, and require dialling a UK phone number for audio. We now have a reliable web desktop sharing system (we have settled on the wonderful Open Source webhuddle, also on SourceForge) and a reliable conference phone system, so there should be no technical issues ;), this time.

If you wish to take part, please click this link for further details....

Some Queries For Data Retrieval From StARlite

It is a public holiday in the UK today (we call these bank holidays, for reasons that seem obscure nowadays). The weather is traditionally bad on such days, and today is no exception, at least the remainder of the week, when we return to work, will be fine and sunny.

We will run a further StARlite schema and query walkthrough webinar shortly, but in the meantime here are some skeleton sql queries, that perform a set of related queries retrieving compounds/bioactivities for a given target. In this case the target is human PDE4A (for which the tid is 3), and human PDE5A (for which the tid is 276). We will walk through getting these unique target identifiers (or tids) on another occasion, but suffice it to say, that this is easy, especially programmatically, using blastp.

Firstly, retrieving a set of potent inhibitors of human PDE4A or PDE5A. There are a number of parameters one needs to set to actually do this (the end-point, the affinity cutoff, etc. Specifically here we have selected high confidence assay to target assignments (the a2t.confidence=7 bit), and where the potency is better than 1000nM for an IC50 measurement. This is a pretty generic query, and piping in the target tid to this covers a surprisingly frequent use the the data.

select  act.molregno, act.activity_type, act.relation as operator, act.standard_value, act.standard_units, 
   td.pref_name, td.organism, 
   a.description as assay_description, 
   docs.journal, docs.year, docs.volume, docs.first_page, docs.pubmed_id, cr.compound_key
from  target_dictionary td, 
   assay2target a2t,    
   assays a, 
   activities act, 
   docs, 
   compound_records cr
where  td.tid in (3,276)
and  td.tid = a2t.tid
and  a2t.confidence = 7
and  a2t.assay_id = a.assay_id
and  a2t.assay_id = act.assay_id
and  act.doc_id = docs.doc_id
and  act.record_id = cr.record_id
and  act.activity_type = 'IC50'
and  act.relation in( '=', '<')
and  act.standard_units = 'nM'
and  act.standard_value <=1000
and  a.assay_type = 'B';

Here is a modified form to retrieve just the compound identifiers (molregno)

select  distinct act.molregno
from  target_dictionary td, 
  assay2target a2t,    
  assays a,  
  activities act
where  td.tid in (3,276)
and  td.tid = a2t.tid
and  a2t.confidence = 7
and  a2t.assay_id = a.assay_id
and  a2t.assay_id = act.assay_id
and  act.activity_type = 'IC50'
and  act.relation in( '=', '<')
and  act.standard_units = 'nM'
and  act.standard_value <=1000
and  a.assay_type = 'B';

Also a common requirement is to get the associated molecule structures from the database - here the syntax is for an sdf format output and the query does not rely on any fancy chemical cartridge manipulation (since we store the molfiles in a clob called molfile in the COMPOUNDS table). The query here simply retrieves the structures, and not the associated bioactivity data. The goofy looking concatenations (||) and newlines (chr(10)) just make sure that a validly formatted sdf file emerges at the end.

select  c.molfile || chr(10) || '> ' ||chr(10)|| c.molregno||chr(10)||chr(10)||'$$$$'||chr(10)
from  compounds c, 
  (select distinct act.molregno
  from  target_dictionary td, 
    assay2target a2t,    
    assays a, 
   activities act
 where  td.tid in (3,276)
 and  td.tid = a2t.tid
 and  a2t.confidence = 7
 and  a2t.assay_id = a.assay_id
 and  a2t.assay_id = act.assay_id
 and  act.activity_type = 'IC50'
 and  act.relation in( '=', '<')
 and  act.standard_units = 'nM'
 and  act.standard_value <=1000
 and  a.assay_type = 'B') t1
where  t1.molregno = c.molregno;