A D V E R T I S E R R E TA I N S S O L E R E S P O N S I B I L I T Y F O R C O N T E N T

A D V E R T I S E M E N T F E A T U R E

B46 | June 2021 | www.nature.com/biopharmdeal

Molecular Health GmbHwww.molecularhealth.com

Molecular Health—intelligent solutionsfor smarter healthcareBetter data for better insights and improved outcomes—Molecular Health has built one of the largest,deepest and fully curated digital platforms for biomedicine in the world. The company’s ‘Dataome’provides the foundation for actionable insights for stakeholders across the healthcare ecosystem.

Molecular Health GmbH is an IT-biotech com-pany developing databases and software solu-tions for precision medicine. Headquartered inHeidelberg, Germany, the company focuses onthe capture, curation, integration, and analysis ofstructured and unstructured biomedical and drugdata sets, to generate molecular understandingof health and disease for physicians, providersand pharma. With its Dataome platform, whichcombines careful quality assessment of the datainput with state-of-the-art artificial intelligence(AI)- and machine learning (ML)-based outputanalytics, Molecular Health offers a unique deepand broad set of capabilities for stakeholders inprecision medicine.

Advances over the past two decades in ‘omics’technologies such as genomics, transcriptomics,proteomics, epigenomics, or metabolomics, andthe association of their readouts with clinicalphenotypes from individual patients, have setthe stage for a new transformative medicine,adding molecular understanding of diseases andtherapeutic interventions to observation-basedmedicine. However, the promise of the ‘omics’revolution and the associated flood of big data hasbeen slower to materialize than initially envisioneddue to the sheer complexity of the large data setsbeing acquired and the analytical challenges theypose. One of the largest hurdles facing the field hasbeen assessing data quality and reproducibility,and simply being able to tell apart validated datafrom unvalidated ones.

Recognizing that addressing this main challengewould be the premise for improving the predic-tive value of any AI/ML–based data-mining plat-form, Molecular Health adopted a two-prongedapproach that puts equal emphasis on ensuring thehighest quality of the data being considered andon developing the most sophisticated analyticalpipelines to extract and deliver actionable insightsof real value for decision makers (Fig. 1).

“As a pioneer in precision medicine, we ensurethat available clinical and molecular knowledgeis made accessible and usable for daily patientcare,” said Friedrich von Bohlen, CEO of MolecularHealth (Fig. 2). “A strategy of paying equal atten-tion to the quality of data feeding data bases andto the strength of analytical tools themselvesmakes us the ideal partner for physicians, provid-ers and pharma seeking to improve outcomes forpatients or to develop better new drugs.”

Dataome–ground supportinglife science’s big dataSince the DNA sequencing of the human genomewas completed almost two decades ago, a newdimension has become actionable for medicine:the molecular dimension, which does not only readthe genetic code, but puts a meaning to molecularinteractions in healthy and disease states.

The complexity of the steadily accumulatinghealth data has been increasing at an unprec-edented pace and it still is. The development ofspecialized ‘omics’ technologies that collect dataat a systems level generates a vast and diverserange of structured and unstructured data typesthat continuously evolve over time. This poses achallenge not only for harmonizing and integratinginformation derived from analyzing different levelsof biological organization, but also for determiningthe quality and validity of individual data points.The lack of uniform standards for data collectionand deposition further complicates the processof ensuring the necessary level of quality of thedata entering the analytical pipeline.

From the beginning, Molecular Health recog-nized the centrality of ensuring that data qualitywas at a level comparable to the data processingpipelines being developed for analyzing those

data, and to meet this challenge, the companylaunched its core platform: Dataome.

Molecular Health’s Dataome is a global digi-tal platform for biomedical and pharmaceuticalinformation and real-world patient data built bycontinuously and rigorously mining, curating, andstructuring billions of molecular and clinical datapoints related to genes, molecular mechanisms,phenotypes, diseases, drugs, and outcomes. Theplatform is designed to process both structuredand unstructured data. Structured data are inte-grated into Dataome using a real-time processingpipeline to normalize and quality control the data.Unstructured data are integrated using MolecularHealth’s proprietary natural language processing(NLP) technology, a combination of rule-basedlinguistic, ML and deep learning models trainedto identify critical biomedical facts from sourcesof unstructured knowledge such as drug labels,patents and peer reviewed literature. This initialML-based data gathering step is followed by a pro-prietary curation pipeline powered by an extensivenetwork of biomedical experts.

“With the development of Dataome, MolecularHealth became a trailblazer in the personalizedmedicine space, offering a unique, high-qualitysource of clinico-molecular knowledge,” said von

Cleaning, standardizationreferencing, expert curation

Transformation, knowledgemodeling, data interpretation

Healthcare decisionsupport technologies

Text mining,automated database capture

DATAOMECAPTURE

DATAOMEKNOWLEDGE

DATAOMEANALYTICS

MHAPPLICATIONS

Prediction of clinical trial success (POS)

Genome-guided clinical decision support

Molecular analysisof drug mechanisms

Enabling in silico drugresearch & development

CURATEDCONTENT

KNO

WLE

DGEMODELING

ARTIFICIAL INTELLIGENCE

Structured +unstructured clinical data

Structured + unstructured

human biology data

Fig. 1 | The Molecular Health knowledge pipeline. Molecular Health has built a fully curated digitalplatform for biomedicine, ‘Dataome’, that puts equal emphasis on using the highest quality data andon developing the most sophisticated analytical pipelines to extract and deliver actionable insights ofreal value for decision makers. The pipeline consists of a comprehensive data capture module, a datacuration process, and a knowledge modeling and interpretation platform that support a portfolio ofhealthcare decision support applications.

A D V E R T I S E R R E TA I N S S O L E R E S P O N S I B I L I T Y F O R C O N T E N T

A D V E R T I S E M E N T F E A T U R E

www.nature.com/biopharmdeal | June 2021 | B47

Bohlen. “Dataome captures and contextualizesglobal data about interventions, outcomes andmolecular mechanisms that can be harnessedto develop any disease model, identify new drugtargets, predict drug repurposing, anticipatepatient drug response, to identify personalizedtreatment options for patients, or to predict drugtoxicity and side effects, based on genomic andmolecular evidence.”

Unlocking the potential of DataomeAs early as during the fourth century BC, Aristotleestablished the centrality of data in the scientificinquiry process. Data allow both induction anddeduction of the basic principles that determinenature’s workings. Data, however, only provideinsights when placed within a framework—atheory, a causal model, some other type of logi-cal construct—that establishes their connectionto some version of reality. In other words, data donot speak for themselves but rather need to beembedded within a narrative that organizes theirconnections and derives meaning in the process.

This is the guiding principle behind MolecularHealth’s efforts toward building cutting-edgeanalytical software solutions that maximize theextraction of ‘meaning’ from life sciences’ big datafor decision-making in the space.

Molecular Health has brought together acomprehensive group of experts in the fields ofmedicine, biology, bioinformatics, drug develop-ment, clinical and translational research, medicalpractice and healthcare IT that is continuouslydeveloping and refining novel analytical pipe-lines to deliver actionable intelligence for thecompany’s customers and collaborators. Usinga broad canvas of bioinformatics, chemoinformat-ics, systems biology, clinical data science, AI andML tools, the team has developed a holistic suiteof software solutions for discovery, evaluation,and implementation of new insights in the healthsciences space:• MH Guide: supporting molecular pathologists,

oncologists and other medical professionalsin the clinical annotation of genetic variants,enabling physicians to make precision oncologydecisions. MH Guide is a tool for identifying themost effective, ineffective or potentially toxicmedications for a cancer patient by analyzingthe molecular profile of the patient’s tumor inthe context of all relevant and related informa-tion contained in Dataome. By mapping thepatient’s specific tumor mutations to the world’sknowledge in the field of oncology, MH Guidecreates a patient-specific report that includesa list of recommended medications and othertherapeutic approaches. MH also includes twocomponents specifically designed for the analy-sis of hereditary diseases (MH Guide/BRCA andMH Guide/Mendel).

• MH Effect: helping regulators and pharmaceuti-cal companies re-label drugs, identify repurpos-ing drugs and predict drug side effects. MHEffect is a tool for the comprehensive clinicaland molecular assessment of drug action andadverse events. By tapping into millions ofpatient records of drug outcomes, includingof emerging safety issues, MH Effect helpsdiscover adverse events associated with mar-keted drugs that might have gone unnoticed butcould be relevant to other drugs by simulatingand analyzing drug interaction, and predictpotential safety issues of drug candidates indevelopment. In turn it can identify the desir-able effects of drugs.

• MH Predict: supporting the pharmaceuticalindustry and investors alike make the rightinvestment decisions by predicting the successof clinical trials. MH Predict is a tool that usesAI and ML algorithms for intelligent simula-tions and analysis of clinical study programswith the goal of mitigating the risks inherentin the process of drug development. By givingusers complete control over these algorithmsand making the process fully accessible, MHPredict exponentially increases the level ofconfidence with which the user will implementany of the actionable outcomes determined bythe underlying AI and ML algorithms.

• MH InSilico: enabling in silico drug researchand development. MH InSilico is a service forthe rapid development of testable molecularhypotheses from unique custom-built causal-molecular disease models. This allows to cre-ate hypotheses for both patient care and drugdevelopment. Through MH InSilico, new drugtargets can be found and validated, clinical trialscan be better planned and stratified, and drugscan be more precisely positioned. Moreover,MH InSilico reveals molecular interactions inpatients that can help to guide individual treat-ment schemes. Thus, MH InSilico blends theknowledge of Dataome with the company’sanalytics and experts in collaboration withcustomers’ experts to reach novel insights incare and drug development.

Based on the power to build causal disease modelsMolecular Health sees the convergence of patientcare and pharma R&D. Disease understanding atthe molecular level allows better treatment andbetter drug development.

Friedrich von Bohlen, CEOMolecular Health GmbHHeidelberg, GermanyTel: +49 6221 43851 0Email: info@molecularhealth.com

CON

TACT

Fig. 2 | Friedrich von Bohlen, CEO of Molecular Health: “Empirical evidence without molecularunderstanding is inferior. Molecular Biology stands for the new dimension in medicine. Our data andtechnologies are validated enablers for this transformation in precision medicine, and causalunderstanding of diseases help providers to treat patients better and pharma develop better drugsmore efficiently for patients worldwide.”

“With the development ofDataome, Molecular

Health became a trailblazer inthe personalized medicinespace, offering a unique, high-quality source of clinico-molecular knowledge

Friedrich von Bohlen, CEO,Molecular Health