Why Big Pharma is not addressing the failure of antidepressants

Changing views on depression Shutterstock

Around a quarter of people experience depression at some point in their lives, two-thirds of whom are women. Each year over 11 million working days are lost in the UK to stress, depression or anxiety and there are more than 6,000 suicides. The impact of depression on individuals, families, society and the economy is enormous.

Frontline therapies usually include medication. All the commonly prescribed antidepressants are based on “the monoamine hypothesis”. This holds that depression is caused by a shortage of serotonin and noradrenaline in the brain. Current antidepressants are designed to increase the levels of these chemicals.

The first generation of antidepressants were developed in the 1950s and a second generation came in the 1980s. Products such as Prozac and Seroxat were hailed as “wonder drugs” when they first came onto the market.

In the roughly 30 years since, these kinds of drugs have come to look tired and jaded. Patents have expired and there are doubts about their efficacy. Some scientists even argue the drugs do more harm than good.

Broken model

There has been no third generation of antidepressants. This is despite there having been moon-landing levels of investment in research. The antidepressant discovery process that gave rise to the earlier drugs is clearly broken. It is also apparent that this process had never worked that well, since the only real improvements over the previous 60 years were a reduction of side-effects.

By the mid-2000s the major pharmaceutical companies started disinvesting in this area. Government funding for basic research into depression and charitable funding followed a similar pattern. In 2010 GSK, AstraZeneca, Pfizer, Merck and Sanofi all announced that they had stopped looking for new antidepressants altogether. Professor David Nutt, the former government drug advisor, declared this to be the “annus horribulus” for psychiatric drug research. The likelihood now is that there will be no new antidepressants for decades.

However, there continues to be an urgent and pressing need for more effective treatments. The question the drug companies now need to ask themselves is, did they fail because the task was impossible, or did they fail simply because they got things wrong? Our view is that there was a systems failure.

Brain ventricles Shutterstock

The monoamine hypothesis was not correct. There is little to no clinical evidence to support the notion that depression is associated with low levels of monoamines and predictions based on it were not borne out by research. The drug discovery process built around it was equally flawed and contained at its very heart a basic logical error that meant it was only ever capable of producing drugs with similar effects to the drugs we already had.

However, these failures still give reason for cautious optimism about the development of new drug treatments because they mean depression can now be looked at in ways that weren’t previously possible.

New ideas

We have proposed a new theory of depression we call the “third ventricle hypothesis”. The ventricles of the brain are a network of interconnected spaces filled with a liquid known as cerebrospinal fluid.

The third ventricle hypothesis links the behavioural features of depression, such as sleep disturbance or disruption of appetites for things like food and sex, with the brain structures that contact this space. Other behavioural features associated with depression include hunched posture, avoidance of eye contact and social withdrawal.

Our hypothesis says that this kind of behaviour developed in response to situations where an individual’s survival depends on them remaining in a social group that has become hostile to them. Behaving in this more defensive way helps reduce the probability of further attack by others. The hypothesis goes on to say that this effect is produced by the explosive release of inflammatory substances known as cytokines into the third ventricle.

The third ventricle hypothesis ties in with clinical evidence showing that this ventricle is enlarged in depressives, that depression is associated with elevated levels of cytokines in blood, and other theories of depression that relate to the release of stress hormones. If developed, it could give new insights into the nature of depression and lead to novel approaches to the development of drugs that are used to treat it.

However, it has not proved possible to take this theory forward because its publication in 2010 coincided with the drug companies pulling out of psychiatric drug research. This disinvestment has impacted on all such work in this area. Guy Goodwin, former head of psychiatry at the University of Oxford has warned of a “generational crisis” in terms of capacity to develop new antidepressants unless the withdrawal of pharmaceutical funding is addressed.

A government review of funding for basic research into depression is now needed to revitalise drug company interest in developing new antidepressants, similar to the recent review of the development of new antibiotics. If new funding structures could be proposed, they might provide hope for the millions affected by depression.

Why you should always shake hands with a robot

Technically I'm wardroid, super-intelligent. But I'll spare your human life for now. Chris Bevan/University of Bath, Author provided

There are many unspoken rules of human interaction, whether that’s whether or not to look them in the eyes, the firmness of the handshake, smiling or words of greeting. Little things like this can lead to big judgements about trustworthiness or social acceptability.

What if we can use this type of behaviour to help humans and robots interact? We’ve been exploring a future where public spaces are inhabited by both people and socially aware robots. They might be autonomous robots, or they might be acting as avatars for people for who are participating by being connected to the robot from afar. In either case, the degree to which any social interactions progress smoothly will depend on the judgements that arise.

The symbolic act of shaking hands at the start of a conversation has already been shown to lead to greater cooperation and trust between individuals who interact one-to-one. While part of this stems from expected social norms, a handshake is also a socially acceptable form of touch. Numerous psychological studies over the years have shown that physical touch, even if only fleeting, can have powerful pro-social effects. For example the Midas Touch effect, where waitresses who briefly touch their customers, even when the customer has no conscious recollection of the touch occurring, receive higher tips.

Robot discussions

We wanted to see if the same effect holds between people and robots, or in this case a robotic avatar standing-in for a person. We set up an experiment in which two participants were invited to assume the role of agents negotiating the sale of a piece of land. Key to proceedings was that the buyer knew important additional information that revealed that more profit could be made from the land sale, and the buyers could choose to exploit this advantage if they wished. We measured the extent of cooperation between the buyer and seller with the final profit split agreed between the buyer and seller, with a 50/50 split being the best possible outcome.

Don’t leave me hanging, bro. Aldebaran Robotics, CC BY-SA

The participants did not meet in person before the experiment. In each of the 60 sessions, one participant performed their role tele-presently, via a computer, using an Aldebaran Robotics Nao humanoid robot as their physical representative. Through Nao’s built-in head camera, microphone and speakers they could see and hear their opposite number, who could hear them but not see them or any image of them.

An equal number of buyers and sellers were in the position of communicating via the robot. Conventional wisdom would suggest that the buyers, negotiating from a position of power while being hidden from view and potentially thousands of miles away, would be more likely to exploit their tactical advantage.

Deal or no deal

To study what effect shaking hands at the outset of the negotiation had on the parties' cooperation, we began an equal number of negotiations with a handshake and without. To do so, the tele-present negotiator extended the robot’s arm using a handheld controller, and they could see the robot’s arm extend and be grasped by their counterpart. Touch sensitive sensors in the robot’s hand transmitted a signal when the robot’s hand was grasped which simultaneously made the controller in the distant negotiator’s hand vibrate, creating a subtle sense of connectedness between the pair.

We found that whether the more powerful negotiator (the buyer) was tele-present or not, the act of shaking hands significantly increased cooperation between the negotiators, resulting in a fairer settlement. In practice, what this meant was that even when the buyer did not reveal their tactical advantage to the seller, they still accepted a settlement figure lower than that they’d have achieved by acting purely in their self-interest.

And despite the fact that the negotiator in person could see nothing of the tele-present negotiator – not their face nor a image of it – the use of a robot as a stand-in didn’t seem to affect the degree of trustworthiness built up through the negotiations, nor the degree the negotiators reported that they’d intentionally tried to mislead the other.

This is interesting because it demonstrates that the act of shaking hands, even via a robot, somewhat offsets the loss of visual cues and information about intentions that would normally be found in the face but which in this case is absent.

It certainly suggests that robot designers need to consider more deeply the effect these gestures have and to enable them in their designs.

Deadly aftershocks like the 7.3 quake in Nepal are rare -- but we must take the risk seriously

A man running while his village is evacuated a day after the 7.3 magnitude aftershock earthquake in Nepal. EPA

The 7.3 magnitude earthquake that hit Nepal on May 12, just weeks after the devastating 7.8 magnitude event, should be classed as an aftershock rather than a second earthquake. Although there are relatively few examples of such big aftershocks in history, the tragic events in Nepal demonstrate that we must always be prepared for them.

The first earthquake that violently shook Kathmandu on April 25 and resulted in more than 8,000 deathswas not predicted, but it hardly came as a surprise to seismologists. Nepal is forged by continental collision between the converging India and Eurasia tectonic plates. Earthquakes here are a well-known by-product of the thickening of the Earth’s crust that occurs in response to enormous compressive stress at the margins of the two tectonic plates as they are squeezed together.

Like recent large and extremely deadly earthquakes – the Tohoku earthquake in 2011, the Haiti earthquake in 2010 and the Sumatra earthquake in 2004 – the immediate focus was on rescue and humanitarian efforts rather than planning for another devastating earthquake that may or may not happen.

Appalling aftershock

Yet that is exactly what happened; a magnitude 7.3 earthquake struck the border region between Nepal and China on May 12, just 150 km east of the first event. Already, the death toll has risen to nearly 100, and is expected to climb further in the coming days.

Most earthquakes result from rapid relative motion along a fault in the Earth’s crust. In general, the larger the earthquake, the larger the rupture area of the fault. For example, the giant 2004 Sumatra earthquake (magnitude 9.2) ruptured over a distance of nearly 1500 km. In the days and weeks which follow such sizeable earthquakes, a large number of smaller earthquakes, known as aftershocks, occur in the vicinity of the rupture area, as the crust adjusts itself to accommodate the sudden change in structure.

Impact of the 2011 Tohoku Earthquake on a road in Fukushima EPA

The rate at which aftershocks occur decays over time following the main event, but they can still be recorded months or even years afterwards. In the case of the April 25 event in Nepal, more than 100 aftershocks have already been detected spanning a zone that extends roughly 150 km east of the main epicentre, which is consistent with estimates of the rupture length and direction of the main shock.

May 12’s event occurred towards the eastern end of this earthquake sequence – and the fact that it is smaller than the first event means that it clearly fits the definition of an aftershock. Ostensibly, a magnitude 7.3 earthquake may seem similar in size than a magnitude 7.8 earthquake, but the logarithmic scale used by magnitude estimates actually means that the energy release is many times smaller, which is reflected in the reduced rupture area of the fault (by a factor of 4-5 in this case). However, size is not the only thing that matters when it comes to the destructive power of earthquakes; the geology of the region, the location of population centres and the robustness of buildings and infrastructure also play a vital role.

Lessons from history

So how common is it for aftershocks to cause major destruction and casualties on a large scale? It is actually relatively uncommon, although there are notable cases, including the Christchurch earthquake in February 2011. On this occasion, a magnitude 6.3 earthquake shook the city of Christchurch in New Zealand’s south island, causing widespread damage and over 180 deaths.

The devastation following the 2011 Christchurch earthquake. AAP/EPA

This event followed the magnitude 7.1 Darfield earthquake of September 2010, which did not directly result in any loss of life. The lack of casualties was due to the quake being centred some 40 km west of Christchurch, which is a less populated area. It was also due to the earthquake striking in the early hours when most people were asleep. Falling masonary from older buildings would have likely resulted in many deaths had the earthquake struck during the day. While many regard this earthquake as the main shock and the 2011 earthquake as an aftershock, some seismologists argue that since they appear to have occurred on separate fault systems, they could both be regarded as main shocks. Either way, the message from New Zealand and Nepal is similar – destructive earthquakes aren’t always islolated in time and space – and smaller earthquakes can kill too.

But given that it is well known that big earthquakes tend to have more intense aftershocks, is there any way that we can mitigate against their effects so that we can avoid a repeat of the fatalities seen in Nepal? Apart from the advice that can be found in many earthquake action plans – avoid buildings with damage that could lead to fire or collapse in an aftershock and use the “drop, cover and hold” method when an aftershock strikes – there is little more that can be done.

Although we know that aftershocks will happen, it is impossible to pinpoint their exact location, magnitude and timing. Our best defence is to ensure that the effected population is educated about earthquake and aftershock hazard, and apply building codes that are properly informed by seismic hazard maps of the region.

Particle physics discovery raises hope for a theory of everything

The CMS detector at the LHC. µµ/Flickr, CC BY-SA

The standard model of particle physics, which describes every particle we know of and how they interact, was given much credence when the Higgs boson was discovered in 2012. Now, measurements of a rare particle-physics decay at the Large Hadron Collider offer further support for the model – but also hints at ways to find out what lies beyond it.

The standard model is cherished by physicists because it can explain most of the fundamental phenomena in nature by referencing just a handful of elementary particles.

These particles include quarks (one of the components of an atom) and electron-like particles called leptons – along with their so-called antiparticles which are identical but have opposite charge. The model also includes the particles that carry forces between them (photons, gluons, W and Z bosons) and the Higgs.

The elementary particles that according to the standard model makes up matter. By HolgerFiedler nach Benutzer:Murphee via Wikimedia Commons, CC BY-SA

The picture this model provides is remarkably complete and precise – given its (relative) simplicity and the huge variety of very different phenomena which it can explain with amazing accuracy.

Even the sun has spots …

But the standard model is far from perfect. For starters, it does not include gravity. Also, the elementary particles it describes so successfully make up just 4% of the matter in the universe. The rest is a mysterious substance dubbed “dark matter” whose composition we still don’t know. This is one of the reasons why scientists doubt that the standard model can be the true theory of everything“.

For some time now, physicists have been in a desperate search for any phenomena which deviate from the predictions of the standard model, as they could provide clues or hints about the nature of physics beyond it. Any such experimental finds could help test the theories that go beyond the standard model. These include supersymmetry – in which there are copies of all the particles – and string theory, which is an attempt to reconcile quantum mechanics and general relativity.

But so far the standard model has been very resilient, successfully able to explain everything that the experimental physicists managed to throw at it.

That might be about to change. Two collaborations of scientists working at the LHC – one using the Compact Muon Solenoid detector and another carrying out the LHC beauty experiment – at the particle physics lab CERN near Geneva measured the decays of so-called B mesons. B mesons are weird particles made up of a specific quark and an antiquark. They looked at two different kinds of particle: a “neutral” B meson and a “strange” B meson.

The enormous CMS detector that managed to measure the tiny muons. CERN

All B mesons are short-lived and decay spontaneously into a bunch of other mesons. But this study specifically looked at the decays of B mesons into pairs of so-called muons, which are heavier versions of electrons, and their antiparticles.

These decays are particularly interesting because their probabilities can be calculated within the standard model with little ambiguity and high precision. From the experimental point of view, the muons are relatively easy to detect and can be measured with high accuracy.

Starting point for a theory of everything

So, according to the standard model, on average about four of every billion strange B-mesons decay into the muon-antimuon pair (instead of into other particles). For the Neutral B-meson this number is even smaller, about one in ten billion. These are very small numbers indeed and explain why past experiments have failed to detect them.

But the new experiments have been able to observe these decays, and to measure their probabilities. They show that while the strange B-meson decays into muons at the same rate that the standard model predicts, the neutral B-meson does so about four times more often than predicted (although the accuracy here was somewhat lower).

This is a significant new development, as various theories that go beyond the standard model predict larger decay probabilities. These results will help eliminate some of theoretical possibilities for physics beyond the standard model. Knowing that is essential to one day devise the next theory of everything.

The next run of the LHC, which is about to start, should provide the opportunities to improve the precision of these measurements, and to put even more stringent constraints on theories that include physics beyond the standard model – or maybe bring a discovery which does not fit any of the existing ones?

When amateurs do the job of a professional, the result is smart grids secured by dumb crypto

Bright colours, dumb ideas. Oast House Archive, CC BY-SA

Security relies upon good programming and correct adherence to well-designed standards. If the standards are sloppy, then security has been compromised from the outset.

Smart grids, which include the smart meters being rolled out to millions of homes and the upstream equipment used by electricity suppliers, are often secured by the Open Smart Grid Protocol (OSGP), developed by the Energy Service Network Association (ESNA). It’s estimated there are more than 4m devices using OSGP.

If there’s one rule about cryptography it’s that it is difficult to prove there are no weaknesses. Newly developed ciphers and methods are subjected to thorough cryptanalysis and peer review – and it’s not advisable to try and re-invent the wheel and develop a new form of cryptographic method or cipher. And yet the ESNA did just that. Ever since OSGP was standardised in 2012 ESNA has been under fire for its decision, and now researchers have discovered just how bad that decision was.

What is the smart grid?

The smart grid is an internet of devices such as electrical meters and electricity distribution equipment. The idea is that network connectivity provides better monitoring of energy use, locating faults, and no need to send out someone to read the meter. But with this convenience comes the insecurity of being attached to the public internet – hence the need for protection.

Normally these devices communicate using secure tunnels. This shows a secure tunnel created between the power company and the home device.

Internet connected smart grid devices. Bill Buchanan, Author provided

The power company sends its public key to the smart meter, which creates a new session key, encrypts this with the power company’s public key, and passes it back. The power company, using its private key, decrypts this to determine the session key for the connection. Both sides will then use their copies of the session key to encrypt traffic passed between them during the session.

If someone determines the private key of the power company, they can then find out the session key and read – even alter – the communications. The same happened with the Superfish vulnerability, where the private key could be easily determined by trying a few well-known pass phrases.

What’s the weakness?

The current problem with OSGP lies in ESNA’s decision to cook up its own, flawed, cryptographic methods and its non-standard implementation of the RC4 cipher – rather than using any of the well-defined, well-designed cryptography standards that are available.

This vulnerability makes it easy to acquire private keys, something highlighted by academic researchers Philipp Jovanovic and Samuel Neves, who demonstrated how easy it was to crack OSGP’s encryption using easy-to-implement key-recovery attacks.

Their focus was on the OMA digest, which is the core of the authentication infrastructure. A digest is a means of turning data into a cryptographic fingerprint, known as a hash, which is encrypted (“signed”) using the secret, private key. There are many well-defined methods for this, such as HMAC-SHA256 and AES-GMAC, which use standard crytographic methods to produce a signed hash signature.

However, OSGP uses a combination of the OMA digest, the EN 14908 algorithm, and the RC4 cipher. The choice of RC4 seems strange, especially as it has known key- and plaintext-recovery attacks, but the home-brew OMA digest leaves the OSGP with security so weak that the researchers were able to recover private keys using just 13 queries.

We need better locks

For something as important as our energy infrastructure, where the tenth decimal point can mean a cost of millions and where a large-scale outage could lead to serious economic losses, it’s just incredible that ESNA has decided to go it alone and subsequently made a hash of it (if you’ll excuse the pun).

OSGP is currently used in over 4m smart grid devices, which can now be seen as having little in the way of real security. As we scale-up the Internet of Things, there’s a quite reasonable concern that too little thought has been given to how they will be secured.

Also, I think the public key infastructure we have created for the internet is deeply flawed, especially in the cryptographic methods used, many of which are past their useful life. While onion routing, as exemplified by Tor, often gets a bad press because of its use for nefarious activities in the deep web, it’s methods are well-proven and secure.

We really need to start kicking the tyres of our internet infrastructure, pension off those aspects that are past their use-by date and introduce better, newer methods. The more that our economy goes online, the more is at stake. I can’t see someone wishing to patch millions of smart meters or devices as new vulnerabilities are found, but can certainly imagine a load of rogue actors who’d take advantage of them.

This needs to be right, right from the outset. After all, there’s no greater threat to the internet than no electricity to power it.

This is the age of the brain – but bending beliefs and feelings raises political questions

Where did I leave my skull cap? Shutterstock

When it comes to developing ways to enhance human beings, we are increasingly fascinated by all things neurological. If the 20th century was all about the gene, the 21st is shaping up to be the century of the brain. This fascination has even produced a dedicated discipline of neuroethics, which includes the study of the moral case for using medicine to make changes to our personalities, feelings and beliefs.

Society is already used to the idea of mood-altering drugs, both in the form of prescription medicines such anti-depressants and illicit substances such as MDMA (ecstacy). Using both human and non-human subjects, researchers are now beginning to sketch a neurological picture of the brain’s electrochemical functions.

For example, studies using functional magnetic resonance imaging (fMRI) have revealed that particular areas of the brain are associated with particular cognitive events such as our moral emotions and ethical reasoning. Other research has shown that particular chemicals in the brain play an important role in forming relationships and building trust. As a result, scientists and philosophers have begun to imagine ways in which our brains could be altered and perceptions or actions changed as a result.

Research suggests we may be able to maintain our intimate relationships through the use of so-called love drugs. We could use substances that predispose us towards feelings, judgements and behaviours that make us more social. We might even improve our ethical thinking by taking drugs that enhance our cognitive abilities more generally.

Political implications

For the most part, the academic debate suggests that, if neuroscientific research proceeds as imagined, then it would be ethical for us to pursue and embrace this kind of neurotechnological enhancement. But most neuroethicists have neglected to fully consider the political, as opposed to simply individual ethical, implications of such future technologies.

For example, some have argued that morally enhancing humanity is not just welcome but required if we are to survive global challenges such as warfare, climate change and over-population that potentially threaten our existence.

More recently, scientists have been experimenting with a technology that creates an explicitly political use for neurological technologies. Researchers from Leiden University in the Netherlands showed that electrically stimulating a certain part of the brain can be used to reduce an individual’s feelings of prejudice. The authors even suggest that brain stimulation could help us achieve Martin Luther King’s dream of a society in which people will not be judged by the colour of their skin but by the content of their character.

While I do not wish to argue that racism, sexism, homophobia or any other form of prejudice should be accommodated, it is not clear that its neurotechnological “suppression” would, in fact, bring about the society Martin Luther King dreamt of.

After all, how should we judge the character, or even actions, of individuals whose behaviour and perspectives result from interventions in their neurological makeup? Indeed, how should we judge a society whose citizens are not so much subjects as they are “neuro-subjects” - individuals who understand themselves and each other in neurological terms rather than as moral agents?

A broader view

Efforts to enhance human beings tend to focus on the benefits to the individuals concerned. But, whether or not specific neurotechnologies are designed to alter our social behaviour, their introduction and use will certainly have a social impact. In this light, a purely ethical assessment of potential neurotechnologies that change the way we think and feel seems critically incomplete. Any neurotechnology that purports to alter our subjective point of view is essentially political in nature.

For example, debates about love drugs almost entirely ignore sociological research into the changing nature of intimacy under the current capitalist system. Some believe that relationships are now more about the benefits they bring, such as a stable environment for child-raising, mutual development and social belonging, than romantic love. Because it treats love and intimacy as a means to an end, the idea of love drugs could further this social change.

What is required is a more acutely socio-political understanding of not only the neurosciences and what they have to offer but also of neuroethics more generally. Human beings are not simply neurological, or even biological phenomena. We are made up of socio-cultural and historical elements and, like psychological discourses before it, the neurosciences are now part of this realm.

There needs to be a greater level of dialogue and engagement between neuroscience and social science if we are to use the knowledge and technologies that emerge from this domain in a politically, and not just ethically, responsible manner.

Five years of the Tories creates uncertainty for British science

Thousands of people protested against planned cuts to science in 2010. Mark Ramsay/Flickr, CC BY-SA

The election of the new Conservative government generates uncertainty about the future of UK science. While the party’s record in government shows that it recognises the value of research, its commitment to a referendum on EU membership, and lack of commitment to ringfencing spending is worrying. On the other hand, the appointment of the new Science minister may suggest that research will have an increasingly important economic role.

The Conservative manifesto made a number of welcome commitments to research, including plans to “invest in science, back our industrial strategies and make Britain the technology centre of Europe”. It also pledged to continue the former coalition government’s Science and Innovation Strategy, which includes investing £1.1 billion in capital each year until 2010-20. However, it did not commit to protecting non-capital spending on research in the forthcoming spending review.

To understand the changes that might take place, it is important to recognise that continuity, rather than radical change, has been the main feature of UK science policy since David Sainsbury laid the ground work for a cross-party consensus on avoiding the damaging policy flip-flopping of the past.

It is unlikely that the new science and universities minister Jo Johnson, the head of the Number 10 Policy Unit and London mayor Boris Johnson’s brother, will have a radically different view from his predecessor Greg Clark. While not a scientist, Johnson has a reputation for valuing research. If this is maintained, it is likely we will see a continuing, and possibly increased, emphasis on the economic impact of research, as well as additional devolution of powers to cities and regions – both of which could change how and where the money is spent.

Funding and free flow of talent

That said, the commitment to a referendum on membership of the European Union should be a real concern for scientists. The UK is a major scientific player in Europe; 80 of the 302 senior research awards from the European Research Council go to the UK. Research is increasingly international and constraints on the free movement of scientists after an “out vote” would be very damaging.

Just the fact that a referendum is taking place is likely to send a worrying message to the world about the UK’s international commitments. Even incorrect perceptions could influence researchers and students’ willingness to come to the UK, constraining the our ability to recruit the best staff and students. Encouragingly the new science minister recognises the importance of international students to the health of the UK university system and its ability to contribute to society.

The uncertainty surrounding the outcome of the vote may mean firms are tempted to relocate their research and development programmes to guarantee long-term access to EU markets, damaging the wider UK research system. Given these risks, universities and scientists are already active in pressing for a pro-EU vote.

At present a vote to leave looks unlikely – but Cameron may not get much from his renegotiations on the terms of the UK’s membership, as any big changes may require referendums in other member states. But a vote for a Brexit would have a serious negative impact on industrial and academic science as it might cut us off from full participation in EU research networks.

Budgets are also uncertain. The last coalition government protected science against cuts under a so-called “budget ringfence”. The Conservative party’s manifesto has not, however, committed to maintain this arrangement. This is a concern, given the emphasis on economic austerity.

Planned cuts to public spending are probably too severe to be politically viable. Cuts to research spending may therefore be tempting – there is talk in Westminster that the universities have had it relatively easy. This is worrying for science, as the co-ordinated lobbying to protect the research budget by learned societies, grassroots pressure groups and others that existed five years ago, is less visible today.

George Osborne will have to think outside the box to protect science when making cuts to public spending. See Li, CC BY-SA

Even continuing the ringfence is having a damaging effect as inflation takes its toll and causes research councils' budgets to fall by approximately 15% in real terms since 2010. The other major concern is that other areas of spending move inside the ringfence, potentially diluting the overall pot of money available for research. The diagnosis in BIS (not in academic science policy) that the UK’s solid research performance, but weak R&D spending, indicates a problem with commercialisation, may see more money spent there.

Luckily, a number of factors may counter this risk. Science has strong political and public support – and politicians are aware of the cost of the Liberal Democrats' broken promise on student loans. Moreover, politicians increasingly recognise how much science contributes to economic growth. This is important as UK productivity growth has been dire and, while this huge problem was largely ignored in the election, the Treasury is worried. The Science Minister is a close friend of the Chancellor, suggesting policies might be introduced to improve the impact of research on productivity and economic growth.

Finally, the science budget may be saved by a seemingly small-scale change to accountancy rules. As of September 2014, research and development has been moved from “current spending” to “capital investment” in the national accounts. Spending on science now has less impact on the deficit.

Even if funding and the UK’s position in the EU are maintained, there are changes underway in UK innovation policy. The government’s commitment to devolving economic powers to cities and regions is likely to continue given Greg Clark’s promotion to Secretary of State for Communities and Local Government.

Local economic policy making is increasingly extending to cover local innovation policy – and this, in turn will influence research. Universities are probably going to increasing be expected to focus more on generating local economic impact and providing support to local firms. This devolution of economic power may, in the long term, help change the geographic distribution of funding, currently heavily concentrated in London and south-east England. This could be good news for the universities driving the Northern Powerhouse.

So overall, the science system may well be subject to considerable changes. Research is now recognised to be more important when it comes to delivering the productivity increases needed to improve the UK’s economic performance, but this may not be enough to avoid a continuing bumpy ride if significant cuts are made.