Bio-IT World Expo New Product Showcase Voting Opens
September 30, 2020 | Next week the Bio-IT World community will convene for the first Bio-IT World Conference & Expo virtual. While the virtual format is undoubtedly different, the event will offer expanded access to the conference content, plenary presentations, short courses, and awards programs that traditionally fill three days in Boston. Among the content that Bio-IT World is known for: a vibrant exhibit hall featuring the most cutting-edge new technologies, tools, and solutions for the life sciences and bioinformatics space.
Even in the virtual space, the bio-IT community has delivered on innovation. This year's finalists include 33 new products from BC Platforms, Benchling, ConvergeHEALTH, Flywheel, Genedata, Genestack, Globus, GNS Healthcare, Igneous, Illumina, Kanda Software, LabTwin, Lifebit, Linguamatics, NetApp, PercayAI, PetaGene, QIAGEN, Qumulo, Rescale, Riffyn, Synthace, Univa, and Zifo Technologies.
For the first time, we are welcoming the entire Bio-IT World community to join us in judging the new products to name our 2020 Best in Show Award Winners. These awards highlight entries that are innovative, progressive, and will demonstrably improve the bio-IT industry.
Here we’ve outlined the new products (in alphabetical order by company) and included the company’s new product description and technical specifications. Voting is open now, and will close on Thursday, Oct 8 at 5pm.
BC|INSIGHT Cohort Builder from BC Platforms
The BC|INSIGHT Data Warehouse stores all generated data organized in various ways: for example by project, type, and terminology. Using the Cohort Builder, researchers can easily browse and subset by ontology terms, for example search across data to find clinical records with diagnostic codes corresponding to the term “heart.” Clever automated analytics generate summary visualizations to highlight unique data patterns that help maintain perspective across the search. This allows researchers to find specific cohorts within large scale data collections, then further sub-set and build custom cohorts for their specific project purpose. The tool enables users to collect extended data categories—joining in other categories of data associated with the built cohort and enriching the data set. Additionally, users can analyse data and compare across different cohorts. The outcome is that data sets are organised for complex analytics and research applications.
BC|PIPE NGS 1.0 from BC Platforms
The new product features optimize the base product, the fully automated, highly scalable NGS data secondary analysis pipeline called BC|PIPE NGS, for real world use cases. With the new developments, our customers can:
- Run multiple pipelines (e.g. germline, somatic) within our platform, utilizing templates or building custom pipelines. The product allows for multiple pipeline validation, leveraging the same QC framework.
- Eliminate extensive manual intervention and automate the process of alignment and variant calling Exome or Whole genome data through the platform from machine to annotated file for analysis with integrated QC to ensure data quality across runs. Beyond secondary analysis BC|PIPE NGS supports tertiary analysis for variant annotation and interpretation within our app.
- Investigate any hanging or low performing jobs to take actions during execution through advanced performance metrics.
- Modify data analysis logic (WDL support) within the application, run code validation, and then test new logic against the published production pipelines through a new development interface
- Identify areas for performance improvement such as machine quality, sample fidelity, and pipeline efficiency. This information is available with metadata analysis so no individual data needs to be stored, bearing no PHI risk.
Technology wise, this is run in a container infrastructure (Kubernetes for container management), and can run on Premise, in the cloud, or within a hybrid infrastructure allowing architecture fitting each customer’s requirements.
Benchling for Lab Automation Version 1 from Benchling
Benchling for Lab Automation is a vendor-agnostic solution that gives life science R&D companies a centralized, flexible environment to design, document, run, and analyze up to 10,000 samples at once with complete traceability across programs and teams. Key features include robotic programming, large-scale data transfer and transcription, and the ability to track the entire sample lifecycle. This solution easily integrates with almost any laboratory robotics and instrumentation, and Benchling has also partnered with lab automation robotics providers Tecan, Hamilton, and PAA to provide rapid integrations out-of-box through a Plug-and-Play program.
Key new features include:
- Orchestrated robotics: The “Run” feature allows for easy creation of robot programming instructions based on unique, scientist-driven input parameters. Users can also identify and select input samples digitally stored in Benchling Registry.
- Parse and document runs: Pull results and run outputs directly into Benchling. Automatically create and update entities in Registry and volumes of samples and reagents in Inventory.
- Track the entire sample lifecycle: Benchling for Lab Automation seamlessly integrates with liquid handlers, plate readers, imaging instruments, chromatography systems, and other instruments. This provides end-to-end sample traceability (e.g., which mouse, tissue, hybridoma, supernatant results are linked to a specific sub-clone?) throughout workflows without manual intervention.
Benchling Life Sciences R&D Cloud from Benchling
The Life Sciences R&D Cloud consists of (1) a suite of six natively unified, browser-based applications hosted on AWS and built with Python and React and (2) an interconnected, bio-intelligent platform built on Amazon RDS and Postgres.
The Life Sciences R&D Cloud's six applications are Notebook, Molecular Biology, Registry, Inventory, Requests, and Workflows. This suite of applications, natively built for large molecule R&D, empowers scientist productivity, sample tracking, and R&D process management.
The Cloud’s bio-intelligent platform stores all types of R&D data, from DNA and protein sequences, to experimental results and metadata, to any custom objects the user defines such as animals, assays, and even greenhouses. In an industry first, this platform features bio-intelligent linkages that interconnect all R&D data to satisfy the complex experiment tracking needs of large molecule R&D. These bio-intelligent links automatically connect, for example, plasmid sequences to the physical locations of the cell lines that those plasmids generate downstream. As a result, scientists, managers, and executives can back up decisions with complete, interconnected experimental histories.
In addition, the Cloud can be configured for any organization's use case without any code. Since all changes to user permissions, data types, bio-intelligent links, workflow configurations, and more are all accomplished through a user interface, organizations can use the Cloud to drive rapid R&D process iteration.
MyPath for Clinical, v2.0 from ConvergeHEALTH by Deloitte
MyPath for Clinical is a digital patient engagement solution that includes a cloud-based hub, built on AWS, and front-end applications (web, iOS, Android) that facilitate patient data capture and exchange between patients, study team members, and sponsor team members. Services in the MyPath for Clinical platform align to the Fast Healthcare Interoperability Resource (FHIR) standard.
Release v2.0 includes services that support a clinical trial model (Research Study, StudySite, Research Subject) and definitional services that enable programmatic management of patient activity schedules for study protocols. Plan and Activity Definition services allow admins to define study templates that include patient trial activities (medication doses, surveys, connected device readings, pre-appt tasks, etc.). All study-related services are AWS-based microservices, aligning to the microservice architecture of the underlying platform. Study team members enroll patients onto studies, and the trial subjects’ mobile applications are automatically populated with personalized study activities and a personalized visit schedule.
ConvergeHEALTH Miner - Data Asset Explorer, v1.6 from ConvergeHEALTH by Deloitte
ConvergeHEALTH Miner’s Data Asset Explorer (DAE) serves as a catalog for data assets, insights, and evidence in an organization. It enables life sciences organizations to find, subscribe, and govern access to the data assets that are internal as well as external to the organization. The integration between Data Asset Explorer and AWS Data Exchange serves as a one-stop-shop, providing visibility into the organization’s internal and external data assets and govern access to the datasets. It standardizes the contracting process of acquiring a new data set and compresses the time it takes to receive data by eliminating many of the inefficiencies that exist today in how data moves around the ecosystem.
ConvergeHEALTH Miner Data Asset Explorer is a hosted, secure, web-based application to catalog internal and external data assets, search, manage, govern access and understand the breadth and depth of available data assets and insights in the organization.
Data Asset Explorer Release v1.6 includes the AWS Data Exchange, a new service by AWS for buying and selling file-based data through the marketplace. This is an industry-agnostic service and one of the first industries it is targeting is Life Science and Health Care (LSHC). It will transform how life sciences clients procure and receive Real World Data (RWD) sets.
MyPath for Connected Care, v2.0 from ConvergeHEALTH by Deloitte
MPCC is a digital patient engagement solution that includes a cloud-based hub, built on AWS, and front-end applications (web, iOS, Android) that facilitate patient data capture and exchange between patients and care team members. Services in the MPCC platform align to the Fast Healthcare Interoperability Resources (FHIR) standard. Release v2.0 includes services that enable programmatic management of chronic disease management plans. Plan and Activity Definition services allow care team members to define plan templates that include patient activities (medication doses, surveys, fitness activities, etc.) for disease management programs. The plan and activity definition templates are AWS-based microservices, aligning to the microservice architecture of the underlying MPCC platform. Care team members enroll patients onto plans based on their diagnosis, and patients’ digital companion mobile application is populated with personalized activities that align to their plan.
Flywheel for Life Sciences from Flywheel
Flywheel is an enterprise research data management platform helping life sciences organizations digitally transform and streamline R&D workflows. With a unique strength in managing imaging data, Flywheel offers a scalable research-first platform that can:
- Capture imaging and related data from multiple siloed sources including CROs, research institutions and historical clinical trials to answer new questions cost-effectively
- Curate diverse data to common quality, labeling, and image annotation standards to make data searchable and reusable
- Compute complex, resource-intensive data including image analysis and machine learning with the provenance to support regulatory approvals and scientific reproducibility
- Support secure collaboration to solve problems with multi-disciplinary teams internal and external to your organization
Building on success with academic and research partners, in April 2020, Flywheel launched Flywheel for Life Sciences. Bulk Ingest is a feature that allows for import of multimodal data at scale from Federated Research Partners. Processing pipelines automatically share data with preset recipients allowing Data Managers to standardize all data into an accessible, permanent repository. CLI Sync allows data to be synced with local file repositories. AI algorithms detect and de-identify PHI from images, check data completeness, generate quality reports, and describe the content of data, saving weeks of time spent on curation. Viewing data is simple with Longitudinal Views of clinical trial data, enabling patient longitudinal studies. Advanced Search lets users search in real time with complex queries to efficiently assemble new datasets and widen the retrospective analyses. Customizable access roles and audit logs of user actions support regulatory compliance.
Genedata Imagence 2.0 from Genedata AG
Genedata Imagence 2.0 provides full end-to-end automation of HCS image analysis, reducing the time spent on image analysis by an order of magnitude. This latest version has been further optimized for use of distributed computing resources, enabling computing power to be scaled up when needed. Yielding improvements in automated image analysis for HCS, Imagence capabilties include:
- parallelization of image analysis steps for massive time and cost savings; rapid detect and definition of all cellular phenotypes in an HCS;
- efficient generation of training data and on these train Deep Learning networks for subsequent classification of HCS image sets in production assays; and
- precise quantificiation the relevant pharmacology.
The solution reduces time and costs required for phenotypic image analysis by generalization of expert input, producing quality results from a new experiment takes just seconds vs. weeks typically required by manual optimization with existing legacy systems. The solution seamlessly integrates with Genedata Screener for detailed pharmacological assessment and profiling.
Omics Data Manager from Genestack
We are launching a new product, Omics Data Manager, which addresses the three major problems in biological data management described above - (1) data silos, (2) lack of metadata standards, and (3) unclear data relationships - by providing a modular set of tools and services for data FAIR-fication across life sciences R&D enterprises.
1. Users can easily search both data and metadata across different locations via a fast, federated, and integrated search engine. This includes studies, samples, and multi-omics data (e.g. genomics, transcriptomics, and FACS). Search can be performed via a GUI interface with full-text and faceted metadata search, or programmatically, via REST endpoints and R/Python client libraries. In addition, access to multiple public repositories are available, such as GEO and ArrayExpress.
2. Omics Data Manager harmonises metadata and enforces data standards via customisable metadata template, autocompletion, and validation using (public or proprietary) ontologies and controlled vocabularies, as well as expert-driven curation tools.
3. Omics Data Manager tracks provenance of and relationships between entities such as studies, samples, and multi-omics data. Users can retrieve related entities effortlessly and explore the full context of each entity in order to make better-informed decisions on how to analyse, reuse or integrate data.
Globus from Globus
The Globus service, used by thousands of organizations for research data management, now supports protected data, so scientists can easily work with this data and share it securely and appropriately with collaborators. This comprises management of Protected Health Information (PHI) including data regulated by the Health Insurance Portability and Accountability Act (HIPAA), Personally Identifiable Information (PII), and Controlled Unclassified Information (CUI).
With higher assurance levels for managing protected data, scientists can focus on new insights and discoveries, not on compliance and security concerns. As a result, researchers and teams can work together via approved, compliant methods on richer datasets that will enhance their research and accelerate the pace of discovery.
In addition, organizations have the option to enter into a Business Associate Agreement (BAA) with the University of Chicago, for written assurance that the data will be appropriately safeguarded by Globus in environments governed by HIPAA.
Subpop v 2.0 from GNS Healthcare
The Subpop Web Application is a cutting-edge companion to the REFSTM causal AI and simulation platform for discovering causal drivers of clinical outcomes and costs. Most platforms enable users to observe associations in the data, but they may be subject to bias and confounding. With the Subpop Web Application, users harness the capabilities of Bayesian causal inference at scale to overcome these methodological challenges.
Once a model is generated with REFS, it can be imported into the Subpop Web Application for users to visualize the models, explore key findings, and run virtual experiments without the need for programming skills. Causal models can be visualized graphically in the tool as a network of variables, linked by arrows indicating the direction of the causal relationship. Network visualization allows users to identify causal pathways in the data. Once the causal drivers of a clinical or cost outcome have been identified, users can quantify how changing one of those variables would impact outcomes. To facilitate the process of subpopulation discovery, Subpop runs a number of simulations to help users identify top predictive (treatment impacting) and prognostic biomarkers (disease drivers). Equipped with this tool, users across an organization can run virtual experiments unlocking the mechanistic underpinnings of disease, drug treatment response, and subpopulation discovery.
DataProtect from Igneous
Igneous DataProtect is a backup and archive software solution designed for file intensive environments that meets organization’s tough SLAs without breaking the economic bank.
Igneous DataProtect operates as a SaaS offering that integrates with the API on NAS providers such as NetApp, Isilon, Qumulo and Pure and works with any other NFS or SMB/CIFS enabled system. Igneous simplifies the deployment and policy setting through by importing and operating at the namespace - set and forget. DataProtect efficiently moves and compresses data to any cloud provider including “archive” tiers such as AWS Glacier, Azure Archive Blob Storage and Google Archive Coldline. Retrieval is as fast as searching for data and restoring to any NAS location.
Igneous DataProtect is deployed as a stateless VM connected to the Igneous SaaS service. Get started on terabytes of data, scale to exabytes of data protected through a single management interface. Data is securely written to customers' cloud accounts. With a short VM install, customers can immediately begin protecting their data to any cloud tier or local storage.
The release of Igneous DataProtect in October 2019 was a software only, scale out version specifically optimized for putting data into the cloud. In contrast to solutions that require landing in “hot” tiers, Igneous leverages the economics of “cold” to “archive” tiers directly. To accomplish this, Igneous groups small files together to limit cloud transaction costs, decouples expiration from cloud compaction to limit cloud storage costs. Igneous is the economic equalizer to enable cloud backup and archive.
DataDiscover from Igneous
Igneous is architected for high-performance file data management at the scale of today’s modern life sciences company. Igneous DataDiscover provides data visibility in a lightweight SaaS offering for greater understanding of file metadata in an intuitive, extensible interface to enable better research decisions.
DataDiscover operates in a default Heatmap view of all data across multiple NAS systems, petabytes of data, and billions of files. The Heatmap view starts with file aging to make economic decisions on projects so researchers can surgically find and archive old projects in a single click to any cloud or less expensive on-premises device.
Extend the Heatmap view with search terms to filter down specific types of data. Find all the results files or raw data across shares and locations. The Heatmap view dynamically shifts to overlay search across the filesystem to provide a “relevance ranking”
Change query terms to provide reports on how much each user, group, or lab is consuming for capacity planning, grants, and chargeback.
Deployed as a virtual machine, DataDiscover is not a heavy infrastructure project and time to insights is measured in hours not days. DataDiscover is deployed as a small, stateless virtual machine that can be installed in as little as ½ hour with results in minutes to hours.
DataDiscover core heatmap functionality was launched in August 2019, new search and reporting was launched March 2020.
The NextSeq 2000 Sequencing System from Illumina, Inc
The Illumina NextSeq 2000 system offers access to on-board, local, and cloud-based analysis software, giving users the flexibility to analyze data in a manner that meets their needs. The on-board DRAGEN (Dynamic Read Analysis for GENomics) Bio-IT Platform offers an ultra-rapid, accurate solution for secondary analysis. The DRAGEN platform uses optimized, hardware-accelerated algorithms for a wide variety of genomic analysis solutions, including BCL conversion, mapping, alignment, sorting, duplicate marking, and variant calling. New pipelines will be made available for a variety of new and emerging applications. The on-board solution provides access to select DRAGEN informatics pipelines such as the DRAGEN-GATK small variant caller, enabling users to generate results in as little as two hours. DRAGEN informatics use best-in-class pipeline algorithms to help novice and expert users overcome bottlenecks in data analysis and reduce reliance on external informatics experts. Users spend less time and effort running production-level pipelines and can focus more on results.
DRAGEN BioIT Platform from Illumina, Inc.
DRAGEN achieves its speed by utilizing field-programmable gate arrays (FPGAs) to accelerated genomic analysis algorithms, including BCL conversion, mapping and alignment, sorting, duplicate marking and haplotype variant calling. In the last year DRAGEN has greatly improved accuracy for germline and somatic small variants, copy-number and structural variants. In the latest release DRAGEN added RNA fusion detection and UMI read collapsing. Finally DRAGEN added population-level joint calling capabilities that scale to tens of thousands of whole genomes, with a target of merging and joint calling hundreds of thousands of whole genomes before the end of the year.
Trapelo from Kanda Software
Trapelo is the only real-time precision medicine platform that enables everyone involved in cancer care to align their decision-making and reimbursement policies with the most current clinical evidence in molecular oncology.
The platform features configurable practice, lab and payor modules that facilitate real-time, evidence-based decision support that lets oncologists know what to tests to order for which patients and from which labs, and then helps them interpret the results for more informed treatment decisions – all of this in the context of the patient’s insurance policies. This optimizes efficiency while giving oncologists confidence they’ve ordered the most relevant tests based on their patient’s disease and clinical stage.
Ensuring reimbursement policies are aligned with current evidence enables oncologists to make informed testing and precision medicine treatment decisions faster. Trapelo is more than a technology tool; it is a healthcare innovation that enables next-generation cancer care to give patients the best possible chance to beat cancer.
Trapelo is: Collaborative. Patient-centered. Continuously Updated. Actionable. Easy-to-use. Learn more here (https://www.interventioninsights.com/for-oncologists)
LifePod Virtual Caregiver from Kanda Software
LifePod’s Virtual Caregiver service expands the capabilities of popular smart speakers (e.g., Amazon’s Echo) with patent-pending innovations that make LifePod an easy-to-use, 2-way voice service for seniors or chronically-ill patients living at home and their caregivers. Unlike traditional voice assistants, LifePod can be set up and controlled by a remote caregiver using LifePod’s online Portal and Dialog Management System. In addition, LifePod’s unique “initiation” capability – enabling LifePod to speak to a senior without first being “woken” by the user – means the senior doesn’t have to remember a “wake word” or how to phrase commands in order to feel more connected and supported. Instead, LifePod provides proactive, voice-first check-ins and reminders each day, and encourages senior users to access voice services (e.g., music, weather, etc.) to enhance their day, based on routines configured online by their caregivers.
LifePod not only enables automated, 2-way voice control and communications but can also add new skills over time-based on the rich data collected via IOT sensors in the senior’s dwelling. Leveraging machine-learning and AI predictive analytics, LifePod will monitor the user’s condition, daily activities and behavioral trends (e.g., sleep, physical routines, medication, etc.) for anomalies or emergency events that can trigger a dialog with the user to collect more information and contact the caregiver directly, via a phone call or text message, as needed.
For Caregivers: Configure proactive reminders, check-ins and easy access to online services for seniors aging-in-place and chronically-ill patients living at home.
For Seniors: LifePod’s AI-powered Virtual Caregiver offers proactive check-ins, reminders, fall detection and sensor-driven healthcare for remote monitoring, support, and connectedness.
Alerts and Reminders: LifePod also offers caregivers the ability to easily configure emergency alerts that will be sent via text, and provides daily reports, sent via email, that summarize activities for family members and the caregiving team as needed.
In late 2018, LifePod announced a partnership with MobileHelp, a leader in mobile Personal Emergency Response Systems (mPERS) that enables people to summon the help they need — be it a family member or an emergency response service — to their location by simply using their voice. The resulting partnership delivers the first fully-integrated, voice-enhanced, emergency response system and virtual caregiver while also providing a proactive solution for healthcare institutions that wish to add voice to their automated remote patient monitoring and voice-enabled response capabilities.
LabTwin from LabTwin
LabTwin is the world's first voice and AI-powered digital lab assistant. LabTwin works alongside scientists to collect data, connect internal and external information streams, help manage experiments and streamline documentation. Powered by voice recognition and machine learning technology, LabTwin’s smart assistant simplifies data capture, structures valuable data, provides on-demand access to scientific information, guides scientists through interactive protocols and provides suggestions to scientists in real time.
Deploit - Beta from Lifebit
Lifebit is democratizing multi-omics, biomedicine & big data analysis with its AI-powered cloud-based system—Deploit. Deploit enables developers and researchers, no matter their computational or data analysis training level, and their corresponding organizations (ie. startups, SMEs, pharmaceutical companies), to instantly run and scale data analysis in a fast, cost-efficient and reproducible way. Deploit automates the analysis process, learns from the data, and provides actionable insights. Further Deploit enables easy sharing of analyses and seamless collaboration no-matter the size of your organization. As an Individual or a company, starting with Deploit requires zero onboarding and zero configuration... just log in, plug, play!
Lifebit CloudOS from Lifebit Biotech
Lifebit CloudOS democratizes the analysis and understanding of genomics big data; the fully federated end-to-end cloud operating system brings computation and analysis to data, wherever it resides, accelerating research. Users are able to scale quickly while drastically reducing costs and speeding time to insights while harnessing the power of AI – positively transformative for the scientists and researchers who share in the mission to radically change how healthcare is done.
Lifebit CloudOS essentially provides a federated digital research environment that powers collaboration and consortia building. As the first fully federated genomics platform that integrates and accommodates all best practices and full compliance models, Lifebit CloudOS guarantees data privacy and security and is the most streamlined, cost-effective and automated data analyses solution that operates across disparate data, teams and organizations.
Lifebit CloudOS enables users to analyse omics data on-demand through a user friendly GUI or command line interface (CLI) - allowing them to bring any analysis to data and further enrich it with their own private or public data. This cloud-native platform exploits infinite resources by scaling even the largest data analyses seamlessly.
I2E 6.0 from Linguamatics
I2E 6.0 is a powerful Natural Language Processing (NLP) software platform which analyzes vast amounts of text, allowing users to address wide-ranging data extraction and knowledge discovery problems across the drug discovery, development and healthcare delivery spectrum—breaking down data silos and delivering critical insights that positively impact human health and wellbeing.
I2E is an end-to-end solution which ingests structured, semi-structured or unstructured documents; performs natural language processing, terminology matching and normalization of the input data; and generates high-quality structured information ready for curation, visualization and downstream analysis.
I2E can be installed locally or used in the cloud, and connects to Linguamatics Content Store which offers comprehensive datasets such as MEDLINE, Clinicaltrials.gov and full-text patents.
In close partnership with our growing customer base we have improved the I2E 6.0 user experience and increased the power of our NLP technology. The most impactful new feature is the web-based graphical user interface. The interface gives users the ability to access the tool from multiple devices and offers centralized query storage.
The new charting provides users with a built-in way to visualize, dynamically filter and sort, their results. This new feature streamlines findings and gives clear, equal prominence to the charts and the results.
We have also improved handling of multi-lingual documents and the detection, extraction and normalization of numerical, or pattern-based, entities in documents. Users looking to extract drug doses, date ranges, percentages, etc. in German, French and Spanish documents can now do so easily and quickly.
NetApp ONTAP AI Reference Architecture for Healthcare: Diagnostic Imaging, version 2 from NetApp
AI solutions demand high accuracy to train and evaluate model fitness. Accuracy is particularly important for assessing and predicting medical conditions in the healthcare industry. Convolutional neural networks (CNNs), a category of deep learning (DL), have proven to be very effective for image recognition and classification applicable to radiology and diagnostics. Therefore, this type of neural network has seen a significant increase in adoption in recent years.
The ONTAP AI reference architecture is an optimized platform for the development of DL models for medical imaging, such as DNN and CNN, and many other healthcare use cases. With the compute power of the NVIDIA DGX-2 system and the data management capabilities of NetApp ONTAP, ONTAP AI enables a full range of data pipelines that span to the edge, the core, and the cloud.
The models and datasets tested in this solution show that ONTAP AI can easily support the workload requirements for model training with MRI datasets, helping data scientists train models to reach higher accuracy and reduce their time to value. This reference architecture enables complex operations in a shorter computation time for training DL models, adding value to healthcare businesses.
CompBio from PercayAI
Other machine learning technologies require extensive data training and curation, yet only provide answers to very specific questions. CompBio™ is an augmented intelligence platform that rapidly identifies contextual relationships to provide insight across all areas of biomedical science. For bench scientists, this means actionable hypotheses from multiomics data sets in just a few hours.
Starting with an omics or multiomics dataset of entities (e.g., genes, metabolites, proteins), CompBio uses a process called Contextual Language Processing (CLP) to identify relevant and meaningful biological processes or pathways by searching the primary literature (PubMed). CLP recognizes the fact that the meaning of a biological concept may differ with context. Thus, the CLP engine breaks language down to the “atomic” level of words and then reassembles those words as biological concepts, then into processes and pathways in a contextually relevant framework. After processing, the most central biological processes and pathways and how they relate to one another are displayed in an intuitive 3D visualizer.
Designed by computational biologists with over 50 years combined experience in pharma, we recognize that success comes not from replacing scientists, but from synergizing the capabilities of computers and humans. As such, CompBio eliminates the “black box” of most AI technologies and empowers the scientist by revealing exactly which primary sources were used to arrive at each result.
Most importantly, CompBio works. CompBio’s technologies are validated and iterated upon daily via 100+ of the nation’s top laboratories at Washington University in St. Louis Medical Center.
PetaSuite Protect v1.0 from PetaGene
PetaSuite Protect 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. Moreover, this is done in a manner transparent to existing tools and pipelines and integrates with existing on-premises and cloud storage infrastructure.
The state-of-the-art prior to PetaSuite Protect is to grant access to users of genomic data on a whole-file or whole-object basis, which means the person it belongs to might be identifiable. While some file-systems support auditing of accesses by internal users, there is very little visibility into what users are doing with this data. And when granting access to external users, there is typically no visibility at all once the data have been transferred to them.
With PetaSuite Protect, users see regular genomic files. When they access these files, they only see the specific regions that they have permissions to view. PetaSuite Protect gives live information on the use of genomic data by those parties, and the ability to immediately grant or revoke access privileges.
Organizations can allocate GA4GH-defined data management roles. Every user access is logged in a tamper-resistant and easily searchable cryptographic ledger. Not only is user and file information recorded but also details of what application was used for access, and what genomic regions were read. Furthermore, decryption and decompression are performed on the client with a transparent high-performance library, rather than by the server. This ensures high scalability across multiple users.
Ingenuity Pathway Analysis with Analysis Match from QIAGEN Bioinformatics
Ingenuity Pathway Analysis with Analysis Match compares the biological signatures for each dataset a researcher analyzes in IPA against signatures of all their previous analyses and to those from more than 50,000 pre-analyzed comparison datasets curated by the QIAGEN company OmicSoft. These are not gene signatures but represent the higher-level biology: what the expression changes are predicted to mean.
The raw gene expression datasets for IPA analysis are disease-relevant datasets downloaded from public data sources such as GEO, ArrayExpress, TCGA, and LINCS and completely re-processed and re-curated with controlled vocabulary in a reproducible pipeline. From these >300,000 samples (so far), more than 50,000 comparisons have been created representing such types as “disease vs. normal”, “treated vs. untreated” or “tissue 1 vs. tissue 2”.
Each quarter, datasets are imported into IPA, analyzed, and published to IPA users. Biological signatures are automatically created for each analysis, consisting of the top predicted Upstream Regulators, Causal Networks, Canonical Pathways, and Diseases and Functions. Signature matches are computed and used to rank the analyses versus the query analysis. From the table of matching analyses, the user can sort and filter by metadata or scores and create a detailed heatmap to display the underlying reasons why their analysis looked similar (or nearly opposite) to other analyses. Recently added to Analysis Match were signatures based on metabolic pathway activation or inhibition, and many thousands more analyses, including the LINCS compendium of >28,000 analyses of well-defined compound treatments or other experimental manipulation of cell lines.
Qumulo Hybrid Cloud File Storage, 2.11.1 from Qumulo
Qumulo’s hybrid cloud file storage is an ideal solution for storing, managing and accessing genomic sequencing and research imaging data. Qumulo is a modern file storage system that can scale to billions of files and that runs in the data center and the public cloud.
Managing throughput and latency are critical for research imaging. Qumulo provides high performance, even with 3D image files that require random small block reads to access the image the researcher wants to see.
Qumulo real-time visibility and control provide information about what’s happening in the storage system, down to the file level, no matter how many files are in the system. Administrators can apply quotas in real-time, so they’re always in control of how resources are allocated. The capacity explorer and capacity trends tools give IT the information it needs to plan sensibly for the future and not waste money due to overprovisioning.
Organizations that want to move some of their genomic analysis and research imagine workloads to the cloud can use Qumulo for AWS. Qumulo has the highest performance of any cloud offering and is the only file storage system in the cloud with a full set of enterprise features, such as multi-protocol support and real-time visibility.
Qumulo ships software every 2 weeks. This methodology, a tenet of Agile development, allows Qumulo to deliver software as soon as it’s ready instead of waiting for the end of a long multi-month cycle. This means that as new customer and market conditions change, Qumulo can reprioritize when it’s in its, and its customers, best interest.
Rescale ScaleX Platform from Rescale
Rescale ScaleX is a platform for high performance computing in a multi-cloud environment. With more than 300 applications ported and tuned in the life sciences and other areas of science and engineering, Rescale enables users to select an application to run, or specify a more complex workflow, and it recommends where to run across multiple cloud providers and architectures, offering pay per use for hardware and software. The fully-managed turnkey solution offers security, compliance, budget controls so users can truly focus on science and engineering. The 2019 ScaleX platform offers enterprise capabilities, increased security, SOC 2 Type 2, HIPAA, and FEDRamp compliance equivalence, access to the latest CPUs and GPUs available, High Performance Storage and High Performance Desktops.
Riffyn SDE v3.3 from Riffyn
The Riffyn SDE is a cloud-based process data system that empowers R&D organizations with a strong digital backbone. Built around a powerful REST API, the platform boost unprecedented flexibility without sacrificing structure or integrity, making it the ideal data management solution for dynamic RD environments.
Data can be ingested into the SDE through several connection tool and becomes automatically associated with the appropriates steps in the process design using patented hyper-graph technology. This results in a functional process flow diagram and creates a FAIR data and process environment for RD personnel.
The Riffyn SDE transforms integrated heterogeneous data into a structured and analysis-ready data table in seconds. Eliminating the tedious hours scientist currently spend cleaning and normalizing data. By streamlining this process, we help organization focus on innovation and discovery.
Riffyn Scientific Development Environment v3.0 from Riffyn, Inc.
Riffyn Scientific Development Environment is used by R&D organizations such as Merck, Sanofi, Novozymes, Cargill, and others to integrate data from various sources into statistical datasets that can be integrated into data analysis pipelines. The latest version, v3.0, adds security, scalability, and a new user interface for visually designing experiments.
Antha, version 20.01.01 from Synthace
Synthace has developed a hardware-agnostic cloud software platform called Antha, which allows scientists to think about the experimental process and not the equipment. The abstraction from specific hardware instructions and data stitching allows digital experimental workflow definitions to be reusable and protocols easily shared. Our platform focuses on three product aspects:
1. Planning and optimizing of liquid handling actions agnostic of the actual liquid handler being used. An important clarification here is this is not just worklist automation but allows for visual flexible design and reuse of protocols.
2. Data aggregation and alignment from diverse sources: microplate readers, qPCR machines and bioreactors. We combine such data with the sample liquid handling steps already generated by Antha to create a fully contextualised dataset.
3. Enabling 'Design of Experiments' (DoE), a statistical approach that allows observation of interactions between experimental factors. Despite its power, the method is cumbersome to execute without dedicated software, especially in the laboratory on automated liquid handlers where applying DoE requires weeks of planning of complex liquid handling steps. Antha removes this burden, allowing DoE to be conducted routinely.
The latest major product release to Antha incorporated more important features: automatic workflow saving and an optimised UI for better visualisation. As importantly has been our integration with the Dragonfly Robot from SPT Labtech and MANTIS from FORMULATRIX—to push the boundary of physical execution of DoE. The use of Antha reduced time spent on liquid handling steps and data aggregation by 75% and 94% respectively.
Navops Launch 2.0 from Univa
Navops Launch is a hybrid cloud management platform that combines public cloud services and on-premise clusters to cost-effectively meet increasing workload demand. This solution optimizes and balances application placement in hybrid cloud services versus on-premise infrastructure and reduces the risk of uncontrolled spending from migrating HPC applications to cloud, to offer enterprise users predictability, monitoring and control of their cloud expenditures.
Navops Launch 2.0 is the only HPC cloud spend management platform in the market today that ensures complete, automated resource provisioning. Version 2.0, the newest enhancement to Univa’s powerful hybrid HPC cloud management product, helps to simplify the placement of enterprise HPC workloads in the cloud while reducing costs by 30-40 percent. It was built to optimize application placement in hybrid cloud services and reduce the risk of uncontrolled spending from running HPC workloads in the cloud.
With features for rightsizing cloud resources and automating hybrid cloud operations, using real-time cloud, application and workload-related metrics, Navops Launch 2.0 gives enterprise users full control over cloud-spend. Additionally, Navops Launch 2.0 introduces an enhanced, intuitive dashboard that provides full visibility into cloud spend vs budgets, offering deep insight, control and monitoring over HPC cloud consumption, and providing CxOs with the certainty to manage cloud spend for multiple cost-centers, projects, departments, users, applications, and cloud resource types.
Sci-Desk 1.0 from Zifo Technologies Private Limited
In the current technology landscape, the scientific systems are either built from scratch in the cloud or companies do lift and shift into the cloud, Sci-Desk is a cloud-based on-demand product offering comprehensive managed services for scientific & lab computing systems spanning discovery, R&D and manufacturing.