Every company is now a tech company. Christopher Mims, a Technology Columnist of WSJ, attests to this in a Dec. 4, 2018, article. Even traditional businesses like brick-and-mortar stores and manufacturers are operating like tech companies. Like those companies, Biotech startups, solving some of the biggest challenges of this century, have much to gain by working as a tech company. Digitizing biotech is a prerequisite — only a digital biotech™ company can operate and scale like a tech company. It is essential to digitize the data biotech companies produce and consume and so that it can be transformed to automate their repeatable mundane tasks and make intelligent machine-guided decisions.
Technology vs. Conventional Biotech Startup
Unlike tech startups capable of launching a product within months from a garage or dorm room and scale into global companies, their biotech counterparts typically spend a decade of work and billions of dollars to reach the same result. The market's steep barrier of entry is daunting and sometimes even insurmountable for biotech startups. There are many culprits responsible for slower processes and poor scalability—chief among them: lack of quality data in the cloud and instant access to resources in the digital environment.
What is digital biotech™?
Digital biotech™ is the dream of creating a digital replica of a biotech company that offers a tech company's scalability and speed. It is the convergence of digital technologies with biotech R&D and operations to bring new breakthroughs to market faster to save and enhance human lives. The dream of digital biotech™ is to give you access to internal and external data conveniently and immediately and transform it to make your processes more efficient and reproducible.
Digital biotech™ will enable the automated collection of scientific and operational data from different sources. This insightful data will then be stored and organized in a centralized place where it can be easily accessed by humans and machines alike.
Biotech companies with high-quality historical data in the cloud will be able to automate all of their repeatable administrative tasks and make data-driven choices at or away from the bench. With digital biotech™, anyone will be able to investigate resource usage (materials, instruments, lab space), monitor lab activities, identify analytical bottlenecks, check and predict instrument maintenance and calibration needs, or check instrument performance in real-time.
With connected data in digitized biotech companies, CXOs and researchers will be able to get a 360-degree view of their entire R&D programs (in-house and outsourced) so that they may obtain insights instantly. Some of the questions biotech teams will be able to get answered in digital biotech companies are the following.
- What is the status of each project?
- What progress have we made towards our goal?
- How much money did we spend on their drug programs?
- What were the hypotheses and key findings of recently conducted research studies?
- How much money did we spend on outsourcing preclinical research of our drug or vaccine?
- When will we get results from the next-gen sequencing experiments?
- How can we get the most critical results as quickly as possible?
- What were the results of the studies the team completed last week?
- Who are our best-performing partners (CROs, product suppliers, and consultants) based on the internal and external ratings?
What is the need for digital biotech™?
Biotech companies are solving some of this century's biggest problems, but the way they operate, especially biotech startups, is stuck in the 20th century.
Most biotech startups do not have their data digitized in one place. Their data is fragmented in paper notebooks, emails, disconnected word docs, powerpoints, ELNs (electronic lab notebooks), LIMS (laboratory information management system), or other point-based software solutions. This prevents them from using their historical data, preserving institutional knowledge, and applying automation and machine learning to their workflows. This makes their research and processes irreproducible and severely limits their scalability.
Cornerstones of digital biotech™
To unleash the power of your company data and create a true digital biotech™ company. You need to keep three cornerstones in mind.
1. Ensure full data integrity
You cannot realize the digital biotech™ vision without ensuring the high quality of data. Regulatory agencies like the FDA recommend nine key principles as a checklist to provide a high data quality that can stand the test of time and reviews. These principles are popularly known as ALCOA+. According to ALCOA+, all data associated with a biotech product should be attributable, legitimate, contemporaneous, original, accurate and complete, consistent, enduring, and available.
Principle #1: Attributable
The first principle of ALCOA+ can be summarised as: The person who performs a data-related task must be identifiable as the person who performed that task.
Principle #2: Legible
The 'L' of ALCOA+ refers to legibility, which means that data should be readable and understandable — painting a clear picture of the step/event sequence that data has passed through.
Principle #3: Contemporaneous
The 3rd principle, "Contemporaneous," means your data activity should be timestamped with a record of when it took place.
Principle #4: Original
Access to original data means that every originally captured piece of data must be retained rather than replaced or deleted.
Principle #5: Accurate
The second 'A' of ALCOA+ stands for accuracy. This means that data should be inputted, stored, and maintained with precision and validity.
Principle #6: Complete
The sixth ALCOA+ principle is that of complete data.
Complete data is characterized by a trackable audit trail to prove that nothing has been deleted or lost.
Principle #7: Consistent
The 7th principle requires that data should display consistently, wherever it is accessed from within your document management system.
Principle #8: Enduring
The 8th principle is "Enduring," which means that any records and information should be accessible and readable during the entire period in which they might be needed, potentially decades after recording!
Principle #9: Available
The last and perhaps the most important is the availability of your data. Documents and records should be accessible in a readable format to all appropriate personnel responsible for their review or operational processes.
2. Make your Data FAIR
FAIR data means your data is Findable, Accessible, Interoperable, and Reusable. The ultimate goal of FAIR is to ensure the data is human and machine-readable and usable. Below is the explanation of FAIR principles.
The first step in making data usable is to be able to find it. Metadata and data must be easy to find for humans and machines. To make data findable, it is essential to describe the data with rich metadata, assign a globally unique and persistent identifier (e.g., DOI) to (meta)data, and (meta)data are registered or indexed in a searchable resource.
It should be possible for humans and machines to access your data with restrictions, if appropriate. To make data accessible, you need to assign a persistent identifier to retrieve data in your data repository. It's also important to put a system in place that authenticates and authorizes data access.
To unleash the full power of your data, you should be able to combine your research data from different data sources.
To uncover new insights, research data should be easily combined with other datasets, applications, and workflows by humans and computer systems. You can make your data interoperable by using controlled vocabularies, keywords, thesauri, and ontologies where possible.
The data you generate should be reusable to replicate experiments, support future research, and make strategic decisions. To make that happen, it is essential to document everything to support proper data interpretation well. Plus, it is crucial to have provenance information to make clear how, why, and by whom the data was generated and processed.
3. Connect your data from different sources and automate all mundane tasks
To form a digital biotech™, you need the ability to manage and integrate data generated at all stages of your R&D lifecycle, from discovery to real-world use. Once you have high-quality data captured and connected, you can create digital twins of different workflows and automate most mundane tasks.
Having data that is consistent, reliable, and well-connected is the prerequisite to create machine learning models and reap the benefits of predictive analytics.