Read more below

seed finding







Therapeutic area is generally chosen by the client due to a combination of existing franchise and portfolio consideration.

We first to assess whether aims are achievable with the technologies currently available and that optimism has not played a too great a role. The pharmaceutical field is awash with projects whose criteria were so specific from the start that they would almost certainly fail.

By working with clients, an independent picture can be built that combines objective analysis with tertiary views from domain experts.

We also encourage clients to re-consider wider opportunities placing more weight on the likelihood of regulatory success than the simpler managerial issues of reorganising an existing sales force to work in a new therapeutic area.


All companies wish to find a high quality molecule. Similarly, all compounds carry a degree of risk.

Our intention is to look objectively and dispassionately at a project to determine whether the compounds look reasonable.

In addition to standard approaches for assessing the molecule itself, such as objective ADME review, we adopt advanced approaches that consider whether similar molecules have been assayed elsewhere and if so whether this may indicate potential side-effects/off target effects.

For monoclonal antibodies we use the unique, advanced approaches of our advisors to estimate antibody humanness and the risks of anti-antibody response.


The drug target and its molecular network of interacting proteins are a key aspect for determining whether an approach will be successful at a molecular level. As part of this, druggability and tractability are considered.

Assessment combines multiple characteristics such as historical assay data, 3D structure binding site and molecular network stability, as well as broader properties of protein family members.

We also determine more general risks to delivering a successful project.

Such issues might include:

  1. -how well the underlying science surrounding the target is understood.

  2. -what other experiments should be done to provide greater certainty before significant investment is made.

  3. -whether newer technologies such as genetic testing should be included to increase the likelihood of success, even if they restrict market size.



Times change. AstraZeneca move into gout therapy with Ardea acquisition despite no existing franchise

Until recently treatments for gout were relatively modest.

Teijin’s gout drug Febuxostat has turned out to be commercially sucessful, with marketing agreements with both Takeda and Astellas for marketing internationally. Yet it was not developed through constant managerial review but rather the tacit energy of a few committed scientists.

Now Astra-Zeneca are looking at the area even though they have no franchise in this area. Their purchase of Ardea for access to the Phase 3 compound Lesinurad, a selective URAT-1 uric acid reabsorption inhibitor for treating gout, takes AZ into a completely new commercial area. This was clearly not their ideal choice but one that ultimately made business sense.



Druggability was once an almost unheard of word but is now commonly used in drug discovery. It describes the likelihood of a compound binding productively to its desired target.

Assessments of successful drugs with properties that are generally agreed to be good, such as RO5 compliance, show that their corresponding targets also have well defined ‘precise’ clefts in which to bind (even if the cleft requires some induced fit to form)

In contrast, those targets with an excellent biological case that have never delivered high quality drugs tend to have poorly defined binding sites. 

Thrombin and FactorX are both examples of an industry-wide general failure to develop orally available, effective inhibitors and it is now generally acknowledged that the binding sites of these serine proteinases do not fulfill the criteria for small molecule drug discovery


Off Targets represent one of the major causes of late stage failure and unexpected events.

But over the past decade, informatics methods have started to provide foundations for exploring this complicated area.

Our analyses, as well as those of others, unambiguously show that for each drug target there are many more proteins against which a drug might bind and that merely screening close homologues is insufficient.

This is particularly the case for compounds that utilise parts of recurrent clefts, such as ATP interaction sites.

Similar sites may exist in distant homologues (structurally similar proteins) as well as unrelated proteins with similar functions. Our approaches help researchers to introduce early-stage checks into their development process, before the costs of clinical trials are introduced.