Molecular Health: Intelligible and Actionable Data for Healthcare

Follow Molecular Health on :

Molecular Health

Rudolf Caspary, CIO and Thomas Koenig, Head of Marketing & PR, Molecular HealthRudolf Caspary, CIO and Thomas Koenig, Head of Marketing & PR
 
Molecular Health, a German-based company, is one of the leading biomedical companies transforming molecular data and medical knowledge into actionable information to support evidence-based clinical, therapeutic and drug development decisions. Molecular Health combines sophisticated data science with scalable software engineering to make data intelligible for healthcare and actionable for patients and various user groups such as physicians, oncologists, molecular pathologists, and pharma among others. Their cloud-based proprietary biomedicine technology platform Dataome analyzes the molecular and clinical data of individual patients against the world’s medical, biological, and pharmacological knowledge to bring more precise diagnostic, therapeutic, and drug safety decisions.

As a major part of modern medicine deals with drug effects on a molecular level, the knowledge about proteins and protein building genes are important. Molecular Health uses state-of-the-art IT technologies from machine learning to in-memory databases or graph databases to create and relate molecular data and provide end-user applications for various industries with actionable results for immediate clinical use and to improve patient lives.

In an interview with CIO Applications, Rudolf Caspary, CIO, and Thomas Koenig, Head of Marketing and PR of Molecular Health, uncover several interesting insights about their organization’s disruptive solutions, unique value proposition, and roadmap for the future.

What are the generic challenges related to data analytics in healthcare?

Rudolf Caspary: Organizations in their quest to find new drugs spend billions of dollars on running medical trials. However, as understanding of diseases on an individual level becomes more complex, many trials fail, and pharmaceutical companies find it difficult to run new trials and discover new drugs. Considering the exponential cost involved in conducting trials, any shortcut or additional insights to redesign and restrict the trial in the early stage are extremely valuable. Today, organizations are trying to predict and understand medical trials using artificial intelligence and machine learning techniques, but the results are far from satisfactory. A deeper insight is required to know how a drug will work for a particular patient as molecular physiology of a disease is very complex. We believe that our solution MH Predict, which is designed to predict clinical trial success using the latest developments in AI and ML and data from Dataome, will drive a true paradigm shift in drug development, resource and investment allocations, and ultimately patient outcomes.

Today, the knowledge about the drugs for a small number of proteins that are tested is limited, but essential knowledge of actionable clinical drugs exists for many proteins. The hurdles are in maintaining comprehensive and discipline-spanning data and integrating them with manifold ontologies in medicine and biology and to run tests as e.g. Whole Exome Sequencing for DNA and RNA sequencing and, most importantly, interpreting all these data in the context of a particular trial, drug or specific patient disease. Data analytical techniques that are often used in research are available for a handful of people, but a clinical solution needs a professional workflow and sharing features. Furthermore, the information needs to be transparent for all stakeholders, including the patient.

We leverage the latest innovations in data science and engineering to perpetually capture the world’s clinical and molecular data. We also enhance and connect this molecular data to expand the dimensions of analytical possibility in AI and ML space and provide the data to our end-users from various industries through applications running in the cloud. These applications allow data mining and connect labs with clinics for sharing data or combining the work results and domain knowledge of a pathologist and oncologist for clinical or research use.

Please throw some light on Dataome technology and its manifold benefits

Rudolf Caspary: The Dataome technology, built and constructed over a decade, enables the quality-controlled capture, integration, and analysis of clinical and molecular knowledge. The platform has been designed to capture and transform raw clinical and molecular data into actionable intelligence for the healthcare industry. The core technology is comprised of three seamlessly integrated innovations:

1. Dataome Capture controls the perpetual capture, enhancement, and integration of global clinical and molecular data sources. This machine-reading framework enables the targeted extraction of biomedical facts that are fed to a proprietary curation infrastructure for review by biomedical experts. This standardized workflow enables the rapid development of proprietary databases to power specific analytical challenges.

2. Dataome Nucleus, the enormous and ever-expanding database of completely integrated clinico-molecular data and knowledge, encompasses the data of millions of patients mapped to trillions of molecular parameters. Nucleus provides more intelligible data for healthcare and an ontological framework that expands the dimensions of analytical possibility.

3. Dataome Analytics provides cutting-edge analytical solutions for the most critical challenges in healthcare. Blending algorithmic excellence with state-of-the-art software engineering capabilities, our talented team brings together experts in the fields of bioinformatics, AI, machine learning, drug development, clinical, translational research, medical practice, and healthcare IT.

The Dataome platform captures and transforms raw clinical and molecular data into actionable intelligence for the healthcare industry


Over the years, Dataome has become one of the largest globally available resources of clinico-molecular data and knowledge. It also provides a single point of access to the world’s biomedical knowledge through our decision support products:

• MH Predict - Our clinical trial prediction solution is a combination of Dataome Nucleus data and our supervised machine learning capabilities. This helps to predict the Likelihood of Success (LOS) of all and any running medical trials and studies including identification of success/failure factors.

• MH Guide - Identifies safe and effective therapies for cancer patients by analyzing the patient’s molecular profile. It matches detected tumor cell mutations (from any kind of biopsy) to the world’s probably broadest knowledge in the field of oncology and helps medical experts to create a patient-specific clinical report with medication recommendations.

• MH Guide BRCA – Helps medical professionals in identifying and reporting genetic variants for Hereditary Breast and Ovarian Cancer (HBOC) following official genetic instructions. It creates individual patient reports based on NGS data from BRCA1/BRCA2 and 18 other genes associated with HBOC.

• MH-EFFECT™ is a first-in-class data analytics solution designed to detect and molecularly analyse adverse events data. By continuously integrating clinical information from adverse event case reports with Dataome’s reservoir of molecular knowledge, MH-EFFECT™ is used to discover adverse events associated with marketed drugs and predict safety issues for drug candidates.

What are the key differentiating factors that help you stay ahead of the curve?

The healthcare industry, in general, is not very transparent. All our products are designed to bring forth transparency for all players, including the patients. This is one of our key differentiating factors. Easy information access is provided via browsers using our cloud-based products without the additional need of IT in the hospital or the laboratory. In order to provide the necessary speed and power for analytical applications, we use in-memory databases or graph databases to create and relate molecular data with drug and disease information. Such patient data are very precisely interpreted so that doctors, even with the limited time they have in their clinical practice, can use these data for a final diagnosis.

To ensure maximum data privacy and security, we abide by the requirements of HIPAA in the U.S, the GDPR in Europe as well as the Gene Diagnostic law in Germany.

Please highlight a case study that brings out the value proposition of Molecular Health in its entirety.

Thomas Koenig: Daten Capital LLP approached us with a mission to build a data-science, AI-driven technique for allocating capital. Daten Capital is a private limited liability partnership firm established to commercialize data-science and AI-driven approaches in investing. They have obtained an exclusive license to apply MH Predict to the financial markets based on the belief that it will help enable the biopharmaceutical industry to deliver new drugs to the market in a much better, faster, and cheaper way. As of today the algorithm has delivered its promise. As another example we collaborate with the renowned German hospital Charité Berlin, and various health insurers and payers for using MH Guide as IVD in ovarian, peritoneal, fallopian tube, and childhood cancer.

What is Molecular Health’s next big step to advance forward from a product innovation perspective?

Rudolf Caspary: Currently, our business operations with Pharma and Providers are active in Europe, Asia and in the U.S., and we continue to rollout new releases and new products in these markets. Major next steps are the registration of the Guide product portfolio as a medical device in Japan and supporting NGS labs with on-site automatic variant detection out of the box for big Whole Exome Sequencing assays. For all organizations that are developing drugs, our plans are to expand MH Predict to include even deeper insights into the root cause of the predicted trial success or failure. And we will further strengthen and position and MH Effect as truly unique technology to gain and augment instant knowledge on targets, pathways and patient information for drug discovery, drug repurposing and drug development.

Read Also