Connect with us

Science & Technology

GDPR Clock is Ticking for the US Companies as Well: Top 7 Tips to Get Ready

Avatar photo

Published

on

General Data Protection Regulation is about to be applicable as from 25 May 2018. Its long-arm teritorrial reach brings obligations not only to EU establishements, but to US based companies as well. Global connection through internet especially underlines the likelihood of such broad application and it will impact US businesses.One of the prerequisits for safe transfer of data between the EU and US is already accomplished by the EU-US Privacy Shield agreement. The European Commission has considered this agreement as providing adequate guarantees for transfer of data. Under Privacy Shield scheme companies may self-certify and adhere to principles stated therein. Yet, there is still less then 3000 companies in the US participating in the Privacy Shield. But GDPR safeguards have still to be followed. Below, we shall look at some of the most profound aspects of compliance with GDPR for the US (non-EU) based companies.

Data protection officer

Although it is not obligatory pursuant the GDPR, it is advisable that a company appoints a data protection officer (‘DPO’) or designate that role to a specific position in the company. DPOcan also be externally appointed. There may be a single DPO for several companies or several persons designated with DPO role in one company. The position needs not necessarily to follow such a title, but it may be a privacy officer, compliance officer, etc. Such person should possess expert knowledge about the GDPR and data privacy, and may have legal, technical or similar background. GDPR was not specific as to requirements of that person, apart from possesing expert knowledge. Role of DPO is toinform, monitor, advise, the controller, processor or employees, to cooperate with supervisory authority, provide training of staff, help in performing data protection impact assesment.

Data Protection Impact Assesment

The further step that companies affected by the GDPR including US companies should do in order to evaluate the risk of data breach is to perform a data protection impact assesment (‘DPIA’). DPIA is a thorough overview of the processes of the company, and can be done with the help of data protection officer. It may include a form or a template with a series of questions, which have to be answered for each processing activity. DPIA has to be detailed and cover all operations in the company. The function of DPIA is to predict situations in which data breaches may occur, and which include processing of private data. DPIA should contain, pursuant to Article 35 of the GDPR, a systematic description of the envisaged processing operations and the purposes of the processing, an assessment of the necessity and proportionality of the processing operations in relation to the purposes, an assessment of the risks to the rights and freedoms of data subjects referred to in paragraph, the measures envisaged to address the risks, including safeguards and security measures. DPIA is a very useful way of showing compliance and it is also a tool that would help to company at the first place, to have an overview of processing activities and an indication of where a breach could happen.

EU representative

A US company (non-EU based company) has to appoint an EU representative if its businessrelates to offering of goods or services to natural persons in the EU, including even free goods or services, or when processing is related to monitoring of behaviour of data subjects in the EU. Behaviour may include monitoring internet activity of data subjects in order to evaluate or predict her or his personal preferences, behaviors and attitudes. EU representative is not obligatory when the processing is occasional or does not include processing on a large scale of special categories of data such as genetic data, biometric data, data concerning health, ethnic origin, political opinions, etc. and when it is unlikely to result in a risk to the rights and freedoms of natural persons. However, given that the exceptions from the duty of designation of EU representative are pretty vague, in most cases companies whose operations are not neglectable towards persons in the EU would have to appoint a reprsentative. Location of such representative would be in one of the EU Member states where the data subjects are located. Representative should perform its tasks according to the mandate received from the controller or processor, including cooperating with the competent supervisory authorities regarding any action taken to ensure compliance with this Regulation, and he/she is also liable and subject to enforcement in case of non-compliance.

Consent matters

GDPR is overwhelmed with one key word of respect the privacy:consent. If companies wish to process data of natural persons that are in the EU, they must first obtain consent to do that. Consent must be freely given, informed, specific and unambigous.

Freely givenconsent presupposes that data subject must not feel pressured, or urged to consent, or subjected to non-negotiable terms. Consent is not considered as freely given if the data subject has no genuine or free choice.Data subject must not feel reluctant to refuse consent fearing that such refusal will bring detrimental effect to him/her. If the consent is preformulated by the controller, which is usually the case, the language of the consent must be clear and plain and easily understandable for the data subject. Further, if there are several purposes for the processing of certain data, consent must be given for every purpose separately. Consent must be specific and not abstract or vague. Silence, pre-ticked boxes or inactivity is not to be considered as consent under GDPR.

Informed consent means that data subject must know what the consent is for. He/she must be informed about what the consent will bring and there must not be any unknown or undeterminedissues. It is a duty of controller to inform data subject about scope and purpose of consent, and such information must be in clear and plain language. But, one must be careful that, as today in the world of fast moving technologies we face overflow of consentsa person has to give in short period of time, there may be an occurrence of ‘click fatigue []1’, which would result in persons not reading the information about the consent and clicking routinely without any thorough thinking. So, the controllers would have to make, by their technical design, such form of a consent, that would make the person read and understand his or her consent. It could be a combination of yes and no questions, changing of place of ticking boxes, visually appealing text accompanying consent, etc.

Consent must be unambiguous, or clearly given. There must not be space for interpretation whether consent is given for certain purpose or not. As to the form of the consent, it may be by ticking a box, choosing technical settings and similar (Recital 32 GDPR).

Data subject gives his consent for the processing of his personal data. However, companies have to bear in mind that data concept in the EU is broadly understood, and that it includes all personally identifiable information (PII), ranging from obvious data such as name and postal address, to less obvious data, but still PII covered by GDPR, such as IP address [2]. On the other hand the IP address is not that clearly considered as PII in the US. In that regard, the protection in the US must be stricter, obliging US based companies to also apply broader EU standards.

Privacy by design implemented

Privacy by design is a concept which brings together the legal requirements and technical measures. It is a nice and smooth way of incorporating law into technical structure of business. Privacy by design, if applied properly at the outset, shall ensure the compliance with the GDPR requirements. It should point out to principles of data minimisation, where only data which is necesssary should be processed, storage limitation, which would provide for a periodic overview of storage and automatic erasure of data no longer necessary.

One of the ways of showing compliance through the privacy by design is ‘pseudonymisation’. Pseudonymization is, according to GDPR, referred to as the processing of personal data in such a manner that the personal data can no longer be attributed to a specific data subject without the use of additional information. Such additional information must be kept separately, so that it cannot be connected to identified or identifiable natural person.Pseudonymisation is not anonymisation and should not be mixed with it. Anonymisation is a technique which results in irreversible deidentification, and since it completely disables identification it is not subject of data protection under GDPR. Pseudonymisation only reduces the likability of a dataset with the original identity of a data subject, and is accordingly a useful security measure [3].

Binding corporate rules

Binding corporate rules (‘BCR’) include set of principles, procedures andpersonal data protection policies as well as a binding clause adopted by the company and approved by competent supervisory authority. Adopting binding corporate rules is not a simple process but means being on a safe track. It is one of the safeguards envisaged by the GDPR. BCR should include according to Article 47 of the GDPR, the structure and contact details of company, categories of personal data, the type of processing and its purposes, application of general data protection principles (such as purpose limitation, data minimisation, limited storage periods, data quality, data protection by design and by default, legal basis for processing, processing of special categories of personal data, ..), rights of data subjects, the tasks of data protection officer, complaint procedures, mechanisms for reporting to the competent supervisory authority, appropriate data protection training to personnel, indication that BCR are legally binding. BCR should additionally be accompanied with privacy policies, guidelines for employees, data protection audit plan, examples of the training program, description of the internal complaint system, security policy, certification process to make sure that all new IT applications processing data are compliant with BCR, job description of data protection officers or other persons in charge of data protection in the company.

Make your compliance visible

Well, if your company has performed all of the above, it has to make it visible. Companies, that are covered with the GDPR, not only do they have to comply, they have to show that they comply. GDPR puts an obligation on controllers to demonstrate their compliance.

From the first contact with the controller, the website must give the impression of compliance. BCR, privacy policies,DPO contact details must be visible in order that data subject may address him in case of data risk or breach. EU representative’s name and contact must be put forward in order to be accessible by the supervisory authority in the EU. Contact form for data subjects with options for access, right to object, erasure, rectification, restriction, should be there.Organisational chart of the company, flow of data transfer demonstrated by data flow mapp.These are only some of the most imporant features that have to be followed.

Non-compliance is a very costly adventure. The adventure that businesses will try to avoid. With systematic planning and duly analysing the necessity of compliance with GDPR, and with clearly defined processes, US companies can put many benefits for the business and attract and encourage data subjects in the EU to freely entrust their datato them. This is a thorough process, but worth accomplishing.

[1] Article 29 Working Party Guidelines on consent,p. 17

[2] According to judgment of the Court of Justice of the EU of 19 October 2016,in case C 582/14,

[3] Article 29 Data Protection Working Party, Opinion 05/2014 on Anonymisation Techniques adopted on 10 April 2014 p. 3

Continue Reading
Comments

Science & Technology

From rockets to spider silk, young scientists wow the jury – and each other!

Avatar photo

Published

on

The winners of the top prize at this year’s EU Contest for Young Scientists. © European Commission, 2023

The 34th annual edition of an EU contest for teenage researchers wrapped up this past week with participants from Canada, Denmark, Poland and Portugal claiming the top prize. 

By Sofía Manzanaro 

Inês Alves Cerqueira of Portugal just spent five days in Brussels and left with a top EU prize for young scientists. 

But ask 17-year-old Cerqueira what she remembers most about the event, which featured 136 contestants from three dozen countries in Europe and beyond, and the much-coveted award gets hardly any mention. 

No worries 

‘I loved listening to all the projects and having conversations about science without having to worry about people judging me or anything like that,’ she said as the 34th annual EU Contest for Young Scientists (EUCYS) drew to a close in the Belgian capital. 

Worries or not, Cerqueira and the other contestants aged 14 to 20 years were judged by a jury of 22 distinguished scientists and engineers from across Europe as part of the official competition. It featured 85 science projects in the running for first, second and third awards that shared a total of €62 000 in prize money. 

The rewards also include scholarships and visits to institutions such as the European Space Agency, nuclear-research organisation CERN and a forum that brings together eight of the largest research bodies in Europe. 

All the participants had already won first prizes in national science competitions. At EUCYS, four projects won the top prize and received €7 000 each. 

Cerqueira claimed hers with two teammates: Afonso Jorge Soares Nunes and Mário Covas Onofre. The three Portuguese, who come from the northern coastal city of Porto, are exploring the potential of spider silk to treat bone diseases including osteoporosis. 

The EUCYS projects, which ranged from rocket science and chronic-pain drugs to climate demographics and river pollution, were as varied as the backgrounds of the participants, who came from as far away as Canada and South Korea. 

Canadian Elizabeth Chen was another first-prize winner for a project on a cancer therapy. The two other top-award recipients were Maksymilian Gozdur of Poland for an entry on judicial institutions and Martin Stengaard Sørensen of Denmark for an initiative on rocket propulsion systems. 

Bright minds 

‘EUCYS is about rewarding the enthusiasm, passion and curiosity of Europe’s next generation of bright minds finding new solutions to our most pressing challenges,’ said Marc Lemaître, the European Commission’s director-general for research and innovation. 

Eagerness and spirit were on general display at the event. So was camaraderie. 

Noemi Marianna Pia, Pietro Ciceri and Davide Lolla, all 17 year olds from Italy, said they felt themselves winners by having earned spots at EUCYS for a project on sustainable food and described the event as a once-in-a-lifetime chance to mix with fellow young scientists from around the world. 

The three Italians want to develop plant-based alternatives to animal proteins. At their exhibition stand, they talked with contagious excitement about their research while holding dry chickpeas and soybeans. 

Lolla said that, while his pleasures include tucking into a juicy steak, he feels a pressing need to reduce meat consumption to combat climate change and preserve biodiversity. 

Sparkling ideas  

On the other side of the venue, 16-year-old Eleni Makri from Cyprus recalled how a classroom chat about summer plans sparked an idea to use seagrass on many of the island’s beaches to produce fertiliser. 

Her project partner, Themis Themistocleous, eagerly joined the conversation to explain how seagrass can recover phosphate from wastewater. The process involves thermal treatment of the seagrass. 

Themistocleous also expressed pride at having been chosen by Makri as her teammate for the competition. 

‘There were a thousand people, but she chose me!’ he said with a wide grin as Makri playfully shook her head in response. 

Science can also be the outcome of a partnership rather than its trigger. Metka Supej and Brina Poropat of Slovenia were brought together by sports, particularly rowing. 

After years of training on the same team, they decided to research the impact of energy drinks on heart-rate recovery. 

Multiple paths 

As they cheered for one another while preparing to say goodbye, the participants at EUCYS 2023 offered a glimpse of the combination of qualities – personal, intellectual, social and even professional – that turn young people into pioneering researchers. 

Gozdur, the Polish top-prize winner, discovered his passion for judicial matters while working at a law firm. Before that, he wanted to study medicine and even dabbled in the film industry. 

His EUCYS project drew on French and Polish criminal-procedure codes to examine the prospects for “restorative justice” – a central element of which is rehabilitation of the convict. The conclusion reached was that ‘penal populism is not beneficial to any party, especially to the victim’s,’ according to a description

Now 19 years old and a law student in Warsaw, Gozdur said he would like international institutions to take up his work so that it influences ‘real-life’ legal norms in the future. 

‘EUCYS showed me that my idea is actually relevant and that it may help societies,’ he said. ‘I would like to fight more for my project.’ 

For Sørensen, the Danish recipient of the top prize, venturing into rocket science as a teenager was no surprise. From the city of Odense, he began computer programming at the age of 10 and was inspired by his father – an electrical engineer – to look into engineering. 

Now 19 years old, Sørensen is striving in his research to create cheaper rocket engines. His project, entitled “Development of small regeneratively cooled rocket propulsion systems”, demonstrated how small rocket engines can be cooled by using a fuel that is a mixture of ethanol and nitrous oxide. 

Sørensen said he’s unsure what his future path will be while expressing interest in pursuing his rocket research.  

‘I would like to continue working on this project,’ he said. ‘And I would like to do something that matters in the world.’ 

Chen, the top-award winner from Canada, has long had a passion for cancer research. 

From childhood, she became involved in fundraisers for a Canadian cancer association and was puzzled about why significant donations had produced no cure. Now 17 years old and in high school, Chen is seeking a therapy that would avoid the often-considerable side effects of conventional treatments. 

Her project focuses on a novel form of immunotherapy based on “CAR-T cells”, which are genetically altered so they can fight cancer more effectively. 

‘I am really interested in going into university right away and then hopefully getting involved in some cancer research because that is just so interesting to me,’ said Chen, who comes from Edmonton. 

The three Portuguese winners – Cerqueira, Nunes and Onofre – said they have developed a partnership as strong as their spider silk and plan to pursue their research while at university with the hope – one day – of conducting clinical studies. 

Called “SPIDER-BACH2”, their project reflects an awareness that osteoporosis will become a growing health challenge worldwide as people live longer. It aims for in vitro production of bone-building cells known as osteoblasts. 

‘The future is bright for us,’ said Nunes. This article was originally published in Horizon, the EU Research and Innovation Magazine.

Continue Reading

Science & Technology

Space Exploration: The Unification of Past, Present and Future

Avatar photo

Published

on

An accreting SMBH in a fairly local galaxy with very large and extended radio jets. © R. Timmerman; LOFAR & Hubble Space Telescope

The enchanting realm of space exploration continues to unfold new wonders with every passing day, sparking a growing interest among individuals to embark on their own cosmic journeys. While exploring space with the aid of private companies that charge fortunes is a privilege usually reserved for billionaire adventurers, there are occasional exceptions that captivate our attention.

Just a few days ago on 8th September, Virgin Galactic’s third spaceflight set out on a brief mission that seized the spotlight due to some interesting details. Three private explorers, Ken Baxter, Timothy Nash, and Adrian Reynard, two pilots and one instructor, were onboard ‘VSS Unity’. However, the presence of two different and unique passengers added a twist to the journey: fossils of our ancient human ancestors. The fossil remains of two ancient species, two-million-years-old Australopithecus sediba and 250,000 years old Homo naledi,  held in carbon fiber, emblazoned with the South African flag,  were part of the Virgin Galactic’s spacecraft ‘crew’ for a one-hour ride, making them the oldest human species to visit space. Australopithecus sediba’s clavicle (collarbone) and Homo naledi’s thumb bone were chosen for the voyage. Both fossil remains were discovered in the Cradle of Humankind – home to human ancestral remains in South Africa.

The episode undoubtedly prompts questions regarding the underlying reason behind sending these fossil remains into the vast expanse of space in the first place. It profoundly underscores the immense power of symbols, speaking to us in ways words cannot. This voyage was not just a journey through space, but a soulful homage to our ancestors. Their invaluable contributions have sown the seeds of innovation and growth, propelling us to unimaginable heights. Now, as we stretch our hands towards the heavens, we remember them – and in this gesture, we symbolise our eternal gratitude and awe for the path they paved, allowing humanity to quite literally aim for the skies. As Timothy Nash said, ‘It was a moment to contemplate the enterprising spirit of our earliest ancestors, who had embarked on a journey toward exploration and innovation years ago.’

Moreover, the clavicle of the Australopithecus sediba was deliberately chosen given that it was discovered by nine-year-old Mathew Berger, son of Lee Berger, a National Geographic Society explorer, who played a major role in discovering both species and handed over the remains to Timothy Nash for the journey. This story serves as a touching testament to the boundless potential of youth, showing us that even the young can be torchbearers in the realm of science, lighting the path of discovery with their boundless curiosity. The unearthing of Homo naledi in 2013 wasn’t just about finding bones; it was a window into our past. This ancient ancestor, with its apelike shoulders and human-like feet, hands, and brain, wasn’t just a distant relative. They were artists and inventors, leaving behind symbols and tools in their cave homes as a silent testament to their legacy. This led to the discovery of more than 1,500 specimens from one of the biggest excavations in Africa’s history. It wasn’t just about digging up the past; it was about piecing together the jigsaw of our very essence, deepening our understanding of the roots and journey of our kind, especially in the heartland of South Africa. Each discovery, each bone, whispered tales of our shared journey, of beginnings, growth, and the undying spirit of exploration.

For those involved in the venture, the occasion was awe-inspiring as it connected our ancient roots to space exploration. However, not everyone is pleased. The event has sparked criticism from  archaeologists and palaeoanthropologists, many of whom have called it a mere publicity stunt and raised serious concerns over such an act given that it poses risks to the care of the precious fossils. It was further argued that the act was ethically wrong, and lacked  any concrete scientific justifications.

Setting aside this debate, the episode connects chronicles of our past with the boundless potential of humankind’s future. It celebrates the age-old quest for exploration shared across millennia. This journey, captivating in its essence, elevates space exploration to a sacred place where fossils, once cradled by the Earth’s soil, now dance among the stars. Just as with pivotal moments in space history, it is also a compelling cue to states that are currently lagging in this race to timely embrace the possibilities of this frontier. Countries, like Pakistan, should draw inspiration from such milestones to fervently chart their own celestial courses.

Upon their return to South Africa, the relics would be displayed in museums and other institutions, offering a chance to the public to view them and draw inspiration. As we witness the rise of commercial space travel, this unique journey provides glimpses of the multifaceted nature of space exploration – one that prompts us to reflect on our past, engage actively with the present and anticipate the future that awaits us. Something Pakistan’s national poet Allama Iqbal eloquently captured in one his verses, translated as: I see my tomorrow (future) in the mirror of my yesterday (past).

Continue Reading

Science & Technology

Artificial Intelligence and Advances in Chemistry (I)

Avatar photo

Published

on

With the advent of Artificial Intelligence technology in the field of chemistry, traditional methods based on experiments and physical models are gradually being supplemented with data-driven machine learning paradigms. Ever more data representations are developed for computer processing, which are constantly being adapted to statistical models that are primarily generative.

Although engineering, finance and business will greatly benefit from the new algorithms, the advantages do not stem only from algorithms. Large-scale computing has been an integral part of physical science tools for decades, and some recent advances in Artificial Intelligence have begun to change the way scientific discoveries are made.

There is great enthusiasm for the outstanding achievements in physical sciences, such as the use of machine learning to reproduce images of black holes or the contribution of AlphaFold, an AI programme developed by DeepMind (Alphabet/Google) to predict the 3D structure of proteins.

One of the main goals of chemistry is to understand matter, its properties and the changes it can undergo. For example, when looking for new superconductors, vaccines or any other material with the properties we desire, we turn to chemistry.

We traditionally think chemistry as being practised in laboratories with test tubes, Erlenmeyer flasks (generally graduated containers with a flat bottom, a conical body and a cylindrical neck) and gas burners. In recent years, however, it has also benefited from developments in the fields of computer science and quantum mechanics, both of which became important in the mid-20th century. Early applications included the use of computers to solve calculations of formulas based on physics, or simulations of chemical systems (albeit far from perfect) by combining theoretical chemistry with computer programming. That work eventually developed into the subgroup now known as computational chemistry. This field began to develop in the 1970s, and Nobel Prizes in chemistry were awarded in 1998 to Britain’s John A. Pople (for his development of computational methods in quantum chemistry: the Pariser-Parr-Pople method), and in 2013 to Austria’s Martin Karplus, South Africa’s Michael Levitt, and Israel’s Arieh Warshel for the development of multiscale models for complex chemical systems.

Indeed, although computational chemistry has gained increasing recognition in recent decades, it is far less important than laboratory experiments, which are the cornerstone of discovery.

Nevertheless, considering the current advances in Artificial Intelligence, data-centred technologies and ever-increasing amounts of data, we may be witnessing a shift whereby computational methods are used not only to assist laboratory experiments, but also to guide and orient them.

Hence how does Artificial Intelligence achieve this transformation? A particular development is the application of machine learning to materials discovery and molecular design, which are two fundamental problems in chemistry.

In traditional methods the design of molecules is roughly divided into several stages. It is important to note that each stage can take several years and many resources, and success is by no means guaranteed. The phases of chemical discovery are the following: synthesis, isolation and testing, validation, approval, commercialisation and marketing.

The discovery phase is based on theoretical frameworks developed over centuries to guide and orient molecular design. However, when looking for “useful” materials (e.g. petroleum gel [Vaseline], polytetrafluoroethylene [Teflon], penicillin, etc.), we must remember that many of them come from compounds commonly found in nature. Moreover, the usefulness of these compounds is often discovered only at a later stage. In contrast, targeted research is a more time-consuming and resource-intensive undertaking (and even in this case it may be necessary to use known “useful” compounds as a starting point). Just to give you an idea, the pharmacologically active chemical space (i.e. the number of molecules) has been estimated at 1060! Even before the testing and sizing phases, manual research in such a space can be time-consuming and resource-intensive. Hence how can Artificial Intelligence get into this and speed up the discovery of the chemical substance?

First of all, machine learning improves the existing methods of simulating chemical environments. We have already mentioned that computational chemistry enables to partially avoid laboratory experiments. Nevertheless, computational chemistry calculations simulating quantum-mechanical processes are poor in terms of both computational cost and accuracy of chemical simulations.

A central problem in computational chemistry is solving the 1926 equation of physicist Erwin Schrödinger’s (1887-1961). The scientist described the behaviour of an electron orbiting the nucleus as that of a standing wave. He therefore proposed an equation, called the wave equation, with which to represent the wave associated with the electron. In this respect, the equation is for complex molecules, i.e. given the positions of a set of nuclei and the total number of electrons, the properties of interest must be calculated. Exact solutions are only possible for single-electron systems, while for other systems we must rely on “good enough” approximations. Furthermore, many common methods for approximating the Schrödinger equation scale exponentially, thus making forced solutions difficult to solve. Over time, many methods have been developed to speed up calculations without sacrificing precision too much. However, even some “cheaper” methods can cause computational bottlenecks.

A way in which Artificial Intelligence can accelerate these calculations is by combining them with machine learning. Another approach fully ignores the modelling of physical processes by directly mapping molecular representations onto desired properties. Both methods enable chemists to more efficiently examine databases for various properties, such as nuclear charge, ionisation energy, etc.

While faster calculations are an improvement, they do not solve the issue that we are still confined to known compounds, which account for only a small part of the active chemical space. We still have to manually specify the molecules we want to analyse. How can we reverse this paradigm and design an algorithm to search the chemical space and find suitable candidate substances? The answer may lie in applying generative models to molecular discovery problems.

But before addressing this topic, it is worth talking about how to represent chemical structures numerically (and what can be used for generative modelling). Many representations have been developed in recent decades, most of which fall into one of the four following categories: strings, text files, matrices and graphs.

Chemical structures can obviously be represented as matrices. Matrix representations of molecules were initially used to facilitate searches in chemical databases. In the early 2000s, however, a new matrix representation called Extended Connectivity Fingerprint (ECFP) was introduced. In computer science, the fingerprint or fingerprint of a file is an alphanumeric sequence or string of bits of a fixed length that identifies that file with the intrinsic characteristics of the file itself. The ECFP was specifically designed to capture features related to molecular activity and is often considered one of the first characterisations in the attempts to predict molecular properties.

Chemical structure information can also be transferred into a text file, a common output of quantum chemistry calculations. These text files can contain very rich information, but are generally not very useful as input for machine learning models. On the other hand, the string representation encodes a lot of information in its syntax. This makes them particularly suitable for generative modelling, just like text generation. Finally, the graph-based representation is more natural. It not only enables us to encode specific properties of the atom in the node embeddings, but also captures chemical bonds in the edge embeddings. Furthermore, when combined with message exchange, graph-based representation enables us to interpret (and configure) the influence of one node on another node by its neighbours, which reflects the way atoms in a chemical structure interact with each other. These properties make graph-based representations the preferred type of input representation for deep learning models. (1. continued)

Continue Reading

Trending