IDEAYA Biosciences Uses CDD Vault to Speed Drug Discovery
Situation
IDEAYA is a precision medicine oncology company committed to the discovery and development of targeted therapeutics for patient populations selected using molecular diagnostics. IDEAYA's approach integrates capabilities in identifying and validating translational biomarkers with drug discovery to select patient populations most likely to benefit from its targeted therapies.
The company, headquartered in South San Francisco, has a world-class team of leading scientists and advisors with extensive knowledge and expertise in cancer biology, small molecule drug discovery, translational biology, and clinical development. The company’s Scientific Advisory Board consists of academic, and industry thought leaders including members of the National Academy of Sciences and the National Academy of Medicine.
Synthetic Lethality
Since inception, a key tenet of IDEAYA’s research has been to exploit the concept of synthetic lethality, an approach for indirectly targeting oncogenic pathways. Certain genetic alterations in cancer cells enable tumor proliferation but also create vulnerabilities – which the company targets with its therapeutics.
Synthetic lethality was first defined by Calvin Bridges in 1922 based on the observation that certain combinations of gene mutations resulted in cell lethality despite the fact that the single mutations in either gene were viable.
“Synthetic Lethality is an emerging area within precision medicine oncology, presenting an opportunity to deliver the next generation of targeted therapies for patients,” says William R. Sellers, M.D., and member of the IDEAYA Scientific Advisory Board. “Advancements in molecular biology research, including genetic knockdown and editing capabilities, are enabling the identification and validation of novel biological targets. Programs in Synthetic Lethality represent exciting opportunities with the potential to be broadly impactful for patients.”
A Robust Pipeline
IDEAYA has built a robust pipeline of potential first-in-class and/or best-in-class precision medicine therapeutics. IDEAYA has target identification, validation, and drug discovery research expertise, including in synthetic lethality—an emerging class of precision medicine targets.
The company’s most advanced pipeline programs include: darovasertib (IDE196), a clinical stage PKC inhibitor for patients with tumors harboring GNAQ or GNA11 mutations, IDE397, a clinical stage MAT2A inhibitor targeting patients with tumors having MTAP gene deletion, which represents approximately 15% of solid tumors, as well as IDE161, a clinical-stage PARG inhibitor for patients having tumors with homologous recombination deficiency (HRD).
With a focus on creating pharmaceuticals that can save lives, IDEAYA scientists take advantage of leading-edge technology to help reduce time to discovery.
Solution
IDEAYA Biosciences, with its passionate pursuit of bringing life-saving therapeutics to patients, needs to work as efficiently as possible. The company has created its own in-house, proprietary machine learning engine called HARMONY™ to help speed its research efforts.
IDEAYA has also assembled best-in-class tools to help its scientists in their important work, including CDD Vault, AWS Data Lake and AWS SageMaker.
CDD Vault
To help manage the mass of data generated by its scientists, IDEAYA deployed Collaborative Drug Discovery’s CDD Vault, the hosted drug discovery informatics platform that securely manages both internal and external biological and chemical data. CDD Vault includes powerful AI algorithms to speed up research by quickly processing data, moving from theory to experimentation faster. The company’s AI identifies patterns and insights in large datasets, uncovering new drug candidates and biological mechanisms. CDD AI algorithms can be used to train and validate QSAR models, as well as to evolve structures using bioisosteric suggestions. All of this helps researchers to streamline early research stages, allocating resources more efficiently and reducing financial risks. In recent years, IDEAYA has registered more than a million novel compounds into their CDD Vault.
AWS Data Lake S3 and AWS SageMaker
IDEAYA harnesses AI and Machine Learning wherever possible to refine its search for the best candidates to advance for further testing. This includes the use of AWS Data Lake S3 and AWS SageMaker XGBoost for supervised machine learning, and AWS Aurora for relational database management.
Benefits
IDEAYA Biosciences has found a number of benefits since adopting CDD Vault, including:
- Integrating CDD Vault with AWS Data Lake and AWS SageMaker
- Taking advantage of CDD Vault integration with Live Design
- Using AI to slash the time to an IND
Integrating CDD Vault with AWS Data Lake and AWS SageMaker
IDEAYA deals with a lot of data as it searches for interesting properties such as protein folding, energy perturbations, protein dynamics, and SMILES (Simplified Molecular Input Line Entry System).
CDD Vault plays a central and unifying role in securely hosting raw data coming in, as well as high-value findings from work in the lab.
“We use best-in-class machine learning models, and then add our own data model from CDD Vault,” says Michael Kappler, Head of Research Informatics at IDEAYA Biosciences. “We synchronize our CDD Vault with AWS Data Lake S3 to populate AWS Aurora databases. And we connect AWS SageMaker machine learning right on top of the data that comes out of CDD Vault. All of this helps us to predict assays more precisely to speed drug discovery.”
Taking Advantage of CDD Vault Integration with Live Design
Scientists at IDEAYA take advantage of CDD Vault integration with Live Design, which enables advanced computational modeling and analysis to rapidly explore distinctive design options to accelerate the drug development process.
“The beautiful thing about our integration of CDD Vault with Live Design is we have everybody on the Vault in Live Design, which accelerates our DMTA (design, make, test, analyze) cycle,” says Kappler. “They have immediate access from the Vault of the data coming in, and they are doing their DMTA cycle in Live Design. This is also important for structural ideation—considering structures that might not otherwise be thought of.”
