BioIT 2019: The New “Official” Future Involves Data Sharing, Artificial Intelligence, Open Source, and More
The 18th installment of Boston’s Bio-IT World Conference was attended by over 3,400 participants and featured more than 185 exhibiting companies, both established and startups. The conference program kicked off with a great presentation by John Wilbanks, Chief Commons Officer at Sage Bionetworks, who focused in his discussion on Open Science – From Ideology to Methodology.
In his provocative presentation – with Sage Bionetworks ten years in the making – Wilbanks examined that research has seen a massive incrementalism which translates into “good in the micro, but not in the macro” or in other words we do know a lot of details about certain diseases (e.g. Alzheimer’s disease) but we still do not understand how to treat them. Wilbanks emphasized that radical changes are required which the pharma industry is longing for. While there is a long list of “magic bullets” – e.g. microarrays, RNAi, NGS – we do appear with those to only achieve incremental changes. And with the current open science approach not much is changing, especially when it comes to impacts for the patient at the center. Some of the challenges observed are analytical challenges: we think of open science as a suite of shared methods. The field needs to think beyond that, as the crowd collectively is smarter than individual experts, and this requires the adoption of open standards. Ideally, we connect little networks to create a big network that brings together an entire suite of data (e.g. genomics, proteomics, clinical data, etc.), ontologies, data standards, tools, trials, data mining approaches, etc. At Sage open science is Just (for justification) and means:
- Transparency increase is justification
- Replication increases justification, but expensive
- Registration increases justification
- Triangulation increases justification
Wilbanks mentioned the All of Us Research Program which is trying to achieve just that – open science which enables trust via a distributed network of communities – with Sage awarded to provide the protocol (but think of the protocol as a plugin, a piece of software).
Interesting quotes from Wilbanks’ talk:
- “Open science is a series of patterns, such as better leveraging data commons, benchmarking, bringing analysis to the data, leveraging developer culture, and open source.”
- “Open science is not a magic bullet – we learned some important things like the wisdom of the crowd isn’t actually that much better, most bioinformaticians do a decent job of analyzing their data”
- “We need open science to address the danger that science becomes just another sector that is consumed by a stack.”
- “Massive incrementalism means you get grants based on papers, not fishing expeditions. So we build information in tiny little steps. Really good in the micro, but not the macro.”
- “Open Data is one of the only ways to ensure justice in AI and algorithms.”
- “We train our algorithms with our own data and there is bias there.”
- “The importance of creating adequate consent trail is a central theme of open science.”
- “Vannevar Bush had a vision of collaborative science, going beyond the single investigator producing ‘paper models’ already in the 50s. Today, we are living it, for example with Sage Bionetworks.”
Quotes from various other Bio-IT talks:
- Vijay Bulusu (Pfizer):
- “Converting data to insights — while there has been an explosion in machine learning and big data, there has been a fundamental lack of progress and investment in improving the “quality” of data. We need to focus on the right thing, converting data to insights.”
- “In the life sciences, we don’t have a big data problem. We have the “lots of small data” problem.”
- Asked the audience: “If they had $1M would they improve the quality of the data or buy a machine learning platform. Nearly everyone would improve the quality of data.”
- Chris Dagdigian (BioTeam): “We’re still bad at helping intermediate users advance to become experts. Again: mentor ship problem!”
- Anthony Philippakis (Broad Institute): “Genomics is at risk of becoming the sport of kings, where only the wealthy can play.”
- Kees van Bochove (The Hyve): “The biggest challenge we have for AI is getting clean data.”
- Sandy Aronson (Partners Healthcare): “AI and Apps are not the solution in hospitals — it’s the more effective care workflows hat the apps and algorithms enable.”
- Susie Stephens (Pfizer): “Lots of AI pilot programs. There is much more work to be done to begin to operationalize AI and automate the use of AI and related insights throughout the business.”
- Dana Vanderwall (BMS): To successfully achieve digital transformation of the lab, it needs to be automated, self-documenting, and interoperable.”
- Anne Carpenter (Broad Institute): “We need more user friendly interfaces for AI to allow average biologists to use the tools.”
Coinciding with Bio-IT were a number of major announcements as listed below:
Five commercial products were selected and honored among the 31 originally considered new products or product updates with 12 finalists. Winners were selected across four product categories:
- “Where the Puck is Going” Award: Genomenon for Mastermind Reporter – uses AI and ML to accelerate the literature curation [learn more about Genomenon’s Mastermind in the Review of Emerging 2018 Genomic Variants]
- “No BS AI” Award: Genedata AG for Genedata Imagence 1.0 – a deep learning-based solution for phenotypic imaging and corresponding workflows based on convolutional neural networks.
- “Patient-Focused Software” Award: Kanda Software for Trapelo & LifePod Virtual Caregiver – enables alignment of decision making and reimbursement policies with most current clinical evidence in molecular biology
- “Nailed It” Award: PetaGene for PetaSuite Protect v1.0 – enables organizations to manage access to their genomic data by internal and external teams, secured with fine grain regional encryption and deep auditing of data usage.
- And the “People’s Choice Award” went to the faceted browsing and visual analytics ‘2019 Linked Data Ingestion Engine’ created by ONTOFORCE.
New this year was the Innovative Practices Award which highlights outstanding examples of how technology innovations and strategic initiatives can be powerful forces for change in the life sciences, from basic biomedical research to drug development and beyond. Winning entries included:
- Abbvie for 3decision – a web-based structural knowledge management solution (nominated by Discnergine)
- Novo Nordisk for integrating NLP to generate actionable insights from real world data (nominated by Linguamatics)
- Sentieon DNAseq and TNseq software tools for NGS data processing-BWA/GTAK drop-in-replacements [read more about how Sentieon Solves Complex Mathematical Problems – Applied to DNA Sequence Secondary Data Analysis]
- Including an honorable mention of Regeneron Pharmaceuticals and the RGC UK Biobank Consortium for the Data Delivery and Cohort Browser (nominated by DNAnexus)
Following is a review of different announcements timed for release with Bio-IT 2019 and spanning commercial product launches, new partnerships and collaborations, product integrations, and other interesting and relevant topics.
Strategic partnerships and collaborations
- Google Cloud for Life Sciences expands technology partners: BC Platforms (Modular platform from sample data to actionable insights), BrightInsight (for continuous remote patient monitoring for clinical trials) the Broad Institute’s Terra platform (for scalable biomedical research), Genomenon (Public cited variant reference data in BigQuery), Medro (clinical trial acceleration), Parabricks (Accelerated Secondary Genomic analysis software suite), Seven Bridges & UK Biobank, and Schroedinger (enables computational drug and materials design).
- Altos (an Acer subsidiary) and WASAI Introduce New Accelerated Genome Sequencing Solution This joint development effort utilizes high-performance Altos BrainSphere AI computing platform combined with WASAI’s proprietary software to create a solution that is faster for genome sequencing without sacrificing accuracy, while also enabling cost savings to make this crucial technology accessible to more hospitals and research institutions.
- PetaGene Partners with NetApp Alliance for Increased User Performance to improve performance and reduce costs for researchers using large data sets within the genomics field. The collaboration will be particularly useful for those transferring backlog processing to the cloud, scaling sample processing and scaling out a global repository for genomic data.
- WekaIO and PetaGene to Showcase an End-to-End Optimized Genomics Workflow The demonstration features WekaIO Matrix (a parallel file system), PetaGene (for data compression), incorporates Sentieon’s genomics tools (accelerated genomic data analysis), and Western Digital’s ActiveScale (for cloud object storage for long-term storage and archival).
- Spectra Logic and Arcitecta Debut Joint Genomics Data Management Solution
New product announcements
- 1Cellbio announces custom targeted bead program to accelerate next phase in single-cell analysis The new program will supply select customers with its proprietary inDrop hydrogel beads synthesized with custom primers that target user-specified transcripts. These are synthesized with custom primers that target transcripts selected by the client which supports focus on sequencing depth on specific genes of interest.
- Seven Bridges Launches Automation Tools and Services which enables biotechnology and biopharmaceutical companies to increase productivity by bringing a diverse set of users into one environment. Scripts written with the Python Automation Development Kit (ADK) automatically gain concurrency, dependency management, memorization, retries, and execution logs, enabling developers to focus on business logic and ultimately, reduce their lines of code by up to 80%.
- Zifo RnD Solutions Announces Launch of Sci-Desk an on-demand, cloud based, Scientific Application Support-in-a-box. Sci-Desk offers comprehensive managed services for Scientific & lab computing systems spanning Discovery, R&D and manufacturing industries.
- Sutter Health, UPMC team up on precision medicine to improve multiple sclerosis treatments Sutter Health is teaming up with the University of Pittsburgh Medical Center and biomedical informatics and data management company DNAnexus on precision medicine research to develop more personalized treatments for multiple sclerosis. The DNAnexus platform will process the whole exome sequencing data and will link the resulting genetic data with extensive clinical data.
- BC Platforms Selected to Deliver Enterprise Research Platform for Qatar Biobank BC Platforms will provide a Research Collaboration Management System (RCMS) and Researchers’ Portal for Qatar Biobank to support the integration of genomic and phenotypic samples, enable secure and streamlined data access, and optimize resource quality and utilization.
- PerkinElmer Joins Accenture’s Life Sciences Ecosystem to Help Drive Innovation in Drug Discovery and Scientific Research designed to help solution providers, software vendors, and life sciences companies team up more effectively with the implicit goal to accelerate drug discovery and improve patient outcomes.