Bio-IT World Names 2024 Best of Show Finalists, People’s Choice Contenders
By Bio-IT World Staff
April 8, 2024 | Bio-IT World announced the 2024 Best of Show Finalists and People’s Choice Contenders today, highlighting 26 new products from 27 different companies that will be on display at the Bio-IT World Conference and Expo, April 15-17, in Boston. The Best of Show Awards will honor several new products live at the event during the Best of Show Awards Reception beginning at 5:15 on Tuesday, April 16.
The People’s Choice Award is open to all entered companies and will be voted on by the Bio-IT World Attendees. All entered companies have described their new products (below) and will have them on display for attendees to learn about and vote on. Voting will take place via code available at each exhibitor’s booth.
The other Best of Show awards are chosen by a panel of peer judges. These judges have already reviewed the field of entries and have narrowed down their choices to 12 finalists. They will visit these finalist companies during the event and choose their final awardees. The finalists for the judges’ prizes are:
Certara, Certara.AI 1.6.1, Booth 217
Combinatics, CellKb version 2.7, Booth 213
Discngine, 3dpredict, Booth 509
expert.ai, Insight Engine for Life Sciences, Booth 119
MaxisIT Inc, SMART Optimizer and v1.0, Booth 824
Nurocor, Nurocor Clinical Platforom v9.0, Booth 809
Osprey Life Sciences, DQA - Digital Quality Analyst v1, Booth 812
QuartzBio, Enterprise Biomarker Data Management v9.1.0, Booth 707
SciBite, SciBite Search 2.4, release is SciBite Chat, Booth 505
The Hyve, Fairspace and 0.7.31-SNAPSHOT, Booth 301
Velsera, Inc., Pangenotyper, Booth 704
Xybion, CompliancePredictor, Booth 224
Here's the full list of People’s Choice Award finalists and the brief product description and technical specifications each company submitted:
Causaly | Causaly Copilot v1 | Booth 419
https://www.causaly.com/products/copilot-genai-life-sciences
The Causaly Copilot combines the power of our state-of-the art biomedical knowledge graph with cutting-edge generative AI models to create the first production-grade Copilot made for life sciences. The Causaly Copilot was designed with the specific needs of scientists in mind. Its natural language search functionality allows for easy question-answering without the need for specialized query language knowledge, making even complex analyses accessible to all users. The Copilot can synthesize key findings into concise summaries or provide in-depth analyses, depending on the user’s needs. This flexibility ensures that scientists can quickly grasp overviews or dive deep into the scientific details as required. The Copilot also gives scientists a way to sift through the noise and stay informed of new developments by continuously monitoring new data and alerting users to significant changes through helpful summaries of the new evidence. Key features include: -Built-in safeguards and guardrails against hallucinations for accurate and trustworthy results. -References and in-line citations for complete transparency, verifiability, and reproducibility. -Integration with the knowledge graph and other scientific data in our extensive data fabric, including patents, clinical trials, genetic data and more. -Deployable against your private data to unlock proprietary insights and data discoverability.
Certara | Certara.AI 1.6.1 | Booth 217
https://www.certara.com/software/certara-ai-platform/
Certara.AI is a secure, flexible platform for deploying life science specific GPTs across your organizational data. With real-time indexing of your data, Certara.AI ensures access to the most up-to-date content related to your area of interest. The result, users can review, validate, and fine-tune generated responses while providing the reference required for trusted GPT use. This enables a powerful platform for applying GPTs across drug development from early discovery to clinical trials to regulatory submission.
Combinatics | CellKb version 2.7 | Booth 213
https://www.cellkb.com
We provide CellKb, a knowledgebase of author-defined gene signatures for cell types. The latest version of CellKb (version 2.7, Feb 2024) provides 80,142 marker gene sets representing 5,326 cell types across 12 species and has grown by 32% over the last year. CellKb has cell type gene expression patterns from both single-cell and bulk RNA-seq experiments. All gene signatures in the CellKb database are manually collected, curated, and annotated with standard ontologies from the source publication. The annotations describe the cell type, anatomical structure, disease, experimental conditions, and other information such as a link to the raw data, source publication, etc. We maintain the high quality of our data by selecting data from publications after a strict review process based on the availability of raw data, experimental method, number of cells studied, computational methods, valid gene identifiers and associated values. Each gene signature is assigned a reliability score computed by statistical analysis. CellKb provides a web-based interface for cell type annotation and biomarker discovery. Users can annotate gene lists for the clusters in their scRNA-seq or spatial RNA-seq experiments. In the latest version, we had added several new features, such as the ability to use all publications in CellKb as a reference to annotate each cell or spot in single cell or spatial RNA-seq data, ability to compare marker genes between cell types across the tissues or diseases of interest, and the ability to predict the cell types from bulk gene and protein expression patterns.
datavisyn | Aevidence 1.0 | Booth 113
https://www.datavisyn.io/products/aevidence/
The Aevidence user journey starts with a single or few entity search field. User searches for their target or disease of interest and hit the search button. The auto-complete search recognizes the search entity and shows several entity-specific apps that allow users to dive deep into their search terms. Each app sits on top of large amounts of public and potentially in-house data and enables the users to understand and interactively explore different aspects of their searched target, such as subcellular location, literature overview, genome browser, sequence analysis with homologs and orthologs, expression analysis, pathway analysis, bioactivity, GWAS and PheWAS, chemical proteomics, structural bioinformatics, and many more. Each app has a clear focus and is designed in a way that provides the most important information at a glance but allows the user to interactively drive the analysis even further. Aevidence also allows collaborative insight gathering and management. Users can create campaigns and try to answer questions about a certain list of targets, such as Target prioritization for chronic kidney disease. The campaign manager can add a list of potentially interesting targets, and the assigned users can now use Aevidence to answer the campaign questions and collect and share deeper insights.
datavisyn | Ordino 1.0 | Booth 113
https://www.datavisyn.io/products/ordino/
To help simplify and streamline integrated biomedical data visualization and analysis we developed Ordino, a single, web-based platform that enables researchers to easily filter, prioritize, and explore large integrated biomedical datasets – and compare in-house experimental results to data stored in public data resources. Ordino facilitates collaborative science and empowers researchers with a framework to interactively visualize, explore, analyze, and annotate data from multiple sources. Other than general purpose off-the-shelf tools, Ordino does support domain specific data types and visualizations. Ordino offers a novel approach to explore data across its original sources. Data is organized in workbenches and user can transitions from one workbench to another, while the data selection in the first workbench is used to filter or attribute the data in the second. Users can walk a path through their data landscape drilling down their analysis. In order to make users understand which data they are looking at, we employed consistent data entity color schemes to help our users to understand where data was coming from and included help texts and tooltips to guide the users as they performed daily tasks. Ordino was developed exclusively for the biomedical domain. One of the central development objectives of this project was to provide a fully-fledged provenance tracking system. At any point in time, Ordino can reproduce its current state and allow users to store and manage “snapshots” of their application state, either for their own usage (bookmarking) or for collaboration.
Discngine | 3dpredict | Booth 509
https://www.discngine.com/
3dpredict is a cloud-based SaaS platform designed to accelerate the discovery of therapeutic antibodies. Accessible via a web front-end or a REST API, the solution allows scientists to easily rank predictions and identify candidates from millions of sequences. The platform supports various antibody formats, including Fab, scFv, and bispecific antibodies. Simply upload single or multi-FASTA files, and specify prediction parameters such as pH range, salt concentration, as well as the desired number of conformers expected for each pH calculation. Utilizing MOE for structure prediction and ensemble property calculations, 3dpredict delivers results obtained from multiple predictions, ensuring higher quality results. The system is scalable, capable of handling large datasets, and integrates seamlessly with the existing in-house IT ecosystem through the REST API.
expert.ai | expert.ai Insight Engine for Life Sciences | Booth 119
https://www.expert.ai
New to the market is the expert.ai Insight Engine for Life Sciences that harnesses the power of AI to mine and aggregate scientific content from scientific literature, clinical trials, research projects and other diverse sources. Now R&D teams can easily synthesize knowledge across the ‘triangle of data sources’: global clinical trials, proprietary scientific articles and public resources for specific therapies. By focusing on specific conditions, the expert.ai Insight Engine enables R&D teams to extract deep insights from complex biomedical data, including genes, proteins, biomarkers and more. The expert.ai Insight Engine delivers several key benefits for life sciences and pharma teams: Research and Development Efficiency: Helps researchers quickly access and analyze a vast amount of relevant information, speeding up the discovery and development of new drugs and therapies. Competitive Intelligence: Provides access to both public and private research articles, helping pharmaceutical companies stay informed about their competitors' activities, including clinical trials, research findings and emerging trends. Clinical Trial Design Optimization: Supports teams in identifying suitable sites and patient populations for clinical trials, improving recruitment and allowing users to compare design complexity based on inclusion/exclusion criteria and mine schedule of assessment data in protocol documents for site and patient burden assessments. Drug Repurposing: Identifies existing drugs that may be repurposed for new therapeutic indications, potentially saving time and resources in drug development. Personalized Medicine: Analyzes data from clinical trials and research articles together, supporting teams in identifying biomarkers and other factors that can be used to develop personalized treatment approaches for patients.
Genomenon | Genomenon Professional Services for Precision Therapeutics: a combination of Expert Curators, Mastermind 3.0 (AI and Genomic Language Processing) | Booth 621
https://www.genomenon.com/pharma/
Product and Technical Specification Summary With the current version of our Mastermind Platform, Genomenon delivers a comprehensive catalog of more than 9,000 gene-disease relationships across 6,000 genes with more being added each week. These curations, based on ClinGen recommendations, are useful to identify causative genes for the diseases of interest and ultimately accelerate variant curation. Variant level information can be used to develop custom panels and inform patient diagnosis when using whole exome (WES) and whole genome (WGS) strategies. Genomenon is the first to curate the clinical exome at the gene level based on ClinGen guidelines, considered the highest industry standard. Our methodology is unique in the industry, Genomenon took a gene first approach, meaning by starting with the gene and looking for disease associations, we were able to uncover additional gene associations when compared with taking a disease first approach. We make this possible by combining an expert genomic curation team with the power of AI. This allowed us to complete the curation of the clinical exome at the gene level in record time, less than 6 months vs an additional 10+ years based on the current curnation rate by the ClinGen working groups. Genomenon simplifies complex genetic data into actionable insights for precision therapeutic development and patient diagnosis. The company’s solutions include software, data, and custom professional services.
Hammerspace | Hammerspace Release 5.x | Booth 413
https://hammerspace.com
Hammerspace software is designed to provide customers with global high-performance access and policy-based services for data that may be siloed across multi-vendor storage environments, including remote sites and the cloud. Unlike solutions that shuffle file copies between incompatible storage types and sites, Hammerspace creates a vendor-neutral high-performance parallel global file system that bridges silos, sites, and clouds so users everywhere are working on the same datasets, regardless of which storage platform the data is on today or moves to tomorrow. Hammerspace’s recent announcement of Hyperscale NAS is the newest innovation, combining the performance of an HPC-class parallel file system with standards-based ease-of-use and reliability features found in Enterprise NAS platforms. At the extreme scale, Meta has implemented Hammerspace's Hyperscale NAS architecture to feed AI training workloads at 12TB/s with data from 1,000 existing storage nodes to a cluster of 24,000 GPUs. In smaller environments, implementing Hammerspace and using the Hyperscale NAS architecture has enabled customers to more than double the throughput of their existing scale-out NAS clusters. Hammerspace’s Hyperscale NAS leverages pNFS 4.2 with Flex Files layouts that is standard in all modern Linux distributions to dramatically increase performance even on existing storage. No client software is needed on servers, or any alteration or agents on the storage. Users/applications see a standard global file system, accessible via SMB, NFS, or S3 protocols. But behind the scenes, Hammerspace can transparently orchestrate data between storage types to manage the different performance requirements needed during the data lifecycle.
MaxisIT Inc | SMART Optimizer and v1.0 | Booth 824
https://www.maxisit.com
Causal AI engine within SMART Optimizer represents a revolutionary advancement in artificial intelligence, exceeding conventional machine learning by prioritizing cause-and-effect relationships in clinical trials data analysis. This paradigm shift based on Causal AI combined with Generative AI architecture pattern built at enterprise scale for pharmaceutical and life sciences industry companies helps move focus on “next best action” to optimize the outcome. SMART Optimizer allows:
To discover cause-effect relationships in clinical trials data by combining subject matter experts’ best of domain knowledge with data-driven, and AI enabled approaches.
To build and discover Causal Models based on cause-effect relationships and causal graphs, that can robustly predict anomalies, figure out the root cause & recommend next best actions to mitigate those anomalies or risks.
To identify deviations from expected outcomes, protocol adherence issues, and safety concerns with continuous real-time data analysis via Dynamic trial monitoring.
To ensure data integrity with automated anomaly detection via intelligent data management Make better decisions and take next best action with SMART Optimizer by understanding root causes automatically ranked by cause-effect severity to the associated anomalous event or risk; design next best action to intervene; and perform powerful what-if analysis as well as estimate counterfactual scenarios to assess potential intervention.
MemVerge | Memory Machine Cloud Edition v2.5 | Booth 712
https://www.mmcloud.io/
Brief Product Description & Technical Specifications:
The MemVerge Memory Machine Cloud (MMCloud) solution on Amazon Web Services (AWS) makes it easy for researchers to migrate their bioinformatics pipelines from on-premises high-performance computing (HPC) to AWS. It also enables executing long-running bioinformatics pipelines cost-effectively on Amazon EC2 Spot instances through automatic checkpoint and restore. New features include: MMCloud Workflow View: Elevates WaveWatcher, allowing bioinformaticians near-real-time monitoring of omics pipeline execution. With genomics pipelines running for extended periods, this feature is crucial for pinpointing issues and spotting optimization opportunities. SpotSurfer with GPU Support: Transforms accelerated workloads like Parabricks and Alphafold by enabling cost-effective execution on Spot instances. This automation of resource management for premium GPU instances saves time and budget, empowering more scientific exploration. SurfZone for IT and Cloud Admins: A tool that addresses the unpredictability of cloud costs, offering real-time HPC resource cost tracking. Unlike existing solutions that reactively terminate workloads, SurfZone offers two proactive responses for exceeding quotas: job cancellation or checkpointing for later resumption. This ensures workloads can be gracefully paused and resumed, optimizing budget management without disrupting ongoing research.
MemVerge | Memory Machine X 1.0 | Booth 712
https://memverge.com
Memory Machine X is software that supports new CXL memory add-in cards (AICs) and CXL E3.S memory modules which allows memory to scale to 32TB inside a single server. Memory Machine X unlocks the potential of "endless memory" by transparently and intelligently placing data on the right tier of memory (HBM in the GPU, DRAM in DIMMs, and DRAM in the CXL AICs and modules).
Nurocor, Inc. | Nurocor Clinical Platforom v9.0 | Booth 809
http://www.nurocor.com
The Nurocor Clinical Platform is a digital protocol solution, comprised of different applications.
Study Designer: create your study design from the ground up, including trial elements, study schematics, and trial arms. Easily update and maintain study design from one centralized location.
Schedule of Activities: guides study designers through an intuitive process to define the study schedule. The schedule of activities table is a critical component of the digital study protocol which drives many clinical operations processes e.g., specimen management, EDC build.
Study and Protocol Elements: turn your study text documents into structured components for reuse across the platform. Easily create, modify, and maintain your study and protocol endpoints for study design, eligibility, interventions, objectives, and endpoints.
Digital Asset Repository: enables the curation, versioning, governance, publishing, and consumption of industry-defined and customer-specific clinical lifecycle standards.
Lean Protocol: gated workflow process that structures a clinical study protocol as a sequence of stages. This enables previously serial processes to run in parallel with protocol authoring.
Specimen Management (new feature): provides a standards-based and automated way to build efficient and consistent biospecimen plans for your studies. These plans precisely document specimen journeys and support generation of operational artifacts such as lab specifications and site manuals.
Authoring (new feature): automatically uses study design elements stored in the platform to populate your clinical trial documents, bringing them into the digital era. This allows authors and reviewers to collaborate in real time.
Osprey Life Sciences | DQA - Digital Quality Analyst v1 | Booth 812
https://ospreylifesciences.com/digital-quality-analyst/
Osprey's Digital Quality Analyst (DQA) has revolutionized manual document quality procedures with its genuine artificial intelligence solution. This AI excels in proofreading documents, ensuring precision in identifying deviations from established standards. At the heart of innovation and efficiency, DQA's AI Analysis Engine streamlines the review process, automatically detecting and suggesting corrections for discrepancies. Tailored for any organization’s rules, it dramatically enhances the consistency, speed, and accuracy of content compliance reviews. More than just an integral component, the AI engine is the cornerstone of DQA, supporting a secure, scalable framework that accommodates business demands of any size without compromising on quality or compliance. By automating routine checks, DQA not only accelerates quality procedures but also frees teams from performing tedious manual reviews. Allowing teams to focus their energy and creativity towards more complex, strategic tasks that demand critical thinking and collaboration. In essence, DQA transforms teams by making them more efficient and by fostering an environment where human intelligence is directed towards innovation and problem-solving.
Pluto Bio | Pluto | Booth 23
https://pluto.bio
Pluto Bio has created the first unified, connected “canvas” environment where scientific collaboration happens in real-time, flexibly, and with end-to-end traceability and reproducibility. In doing so, we are empowering biologists to perform their own analysis without having to wait weeks for help from a bioinformatician, and enabling drug discovery to happen at an accelerated pace. In Pluto’s cloud-based platform, scientists can upload large, raw data files for sequencing and other assays (e.g. RNA-seq, single cell, epigenetics, proteomics, metabolomics) and execute compute-heavy pipelines to preprocess the data with a few clicks. Then, users can select from a broad catalog of industry-specific analyses, which can be added and customized in an intuitive, biology-focused interface to gain immediate insight and address the company’s specific scientific questions. Importantly, the platform integrates with a company’s existing workflows in R or Python, and enables any custom results generated by a bioinformatician to be saved to the canvas in order to be viewable and editable by biologists, who can leave comments and tag relevant collaborators. Pluto releases new analysis modules on a weekly basis and, in 2024, launched a fully end-to-end single cell RNA-seq analysis experience. Unlike single cell “browsers” that allow for exploration of fully processed data, Pluto’s scRNA-seq experience can be used to upload raw data, perform flexible preprocessing steps, and generate publication-ready plots. Plots can be sent to collaborators with secure Pluto share links or embedded back into electronic lab notebooks and a wide variety of other platforms.
Quantum Corporation | ActiveScale Z200 All-Flash Object Storage | Booth 820
https://www.quantum.com/
Quantum’s ActiveScale Z200 All-Flash Object Storage with Cold Storage combines advanced object storage software, high density all-flash servers, and hyperscale tape technology, providing the most advanced solution for data lakes and storage clouds with unmatched performance, durability, and storage efficiency. The Z200 is a scale-out object storage architecture built for ‘always available’ access and massive scale, efficiently storing billions of objects, from terabytes to exabytes with industry-leading durability. The Z200 delivers up to 5X greater throughput (GB/s) and up to 9X more transactions (obj/s) to support high throughput ingest, genomic processing and analytics, AI pipelines, and NoSQL databases. The Z200 is the industry’s only object storage architected for both active and cold data, uniquely capable as a massive data lake, storage cloud, or archive at up to 80% lower cost than alternatives. Its high throughput ingest and unique ability to process and densely store metadata in flash for hundreds of billions of objects stored in the cold storage tier dramatically reduces footprint and cost. With fast, easy access through S3 Glacier interfaces, objects are easily restored in bulk within minutes (not hours or days) for additional analysis, model recalibration, and re-monetization. Biomedical organizations achieve faster insight, discovery, and innovation:
Conduct high performance analysis of data sets at any scale
Secure in-house control of their digital assets
Confidently preserve and protect data over years and decades
Easily access data through standard S3 interfaces to unlock and enrich its value
Independently scale active (flash-based) and cold (tape-based) data sets
Quantum Corporation | Quantum Myriad v 1.0 | Booth 820
https://www.quantum.com/
Quantum Myriad is an all-flash, scale-out file data storage software platform, built for modern applications such as AI and ML. In November 2023, Myriad became generally available for purchase. Myriad delivers deployment flexibility and consistent high performance for data intensive workloads, like AI, frequently experienced within the life sciences. Due to its modern cloud-native architecture, Myriad is an easy-to-use solution that overcomes the limitations of hardware-centric designs. The software features an all new, highly innovative architecture based on the technical capabilities of NVMe and RDMA connections across even very large systems to deliver consistent low-latency performance at any scale. It introduces inline data services such as data tagging, deduplication and compression, snapshots and clones, and metadata tagging to accelerate AI/ML data processing and post-processing operations. And it uses familiar and proven cloud technologies, like microservices and Kubernetes, to deliver cloud simplicity and intelligent, automatic response to its changing storage environment wherever deployed. Myriad also operates on standard high-volume flash storage servers so IT teams can quickly adopt the latest hardware and storage infrastructure for future needs. No custom hardware is required, unlike other all-flash solutions on the market.
QuartzBio | QuartzBio enterprise Biomarker Data Management v9.1.0 | Booth 707
http://www.quartzbio.com
enterprise Biomarker Data Management (eBDM) | Biomarker Intelligence for the Entire Precision Medicine Lifecycle is a SaaS product, powered by QuartzBio’s AI Biomarker Intelligence Platform, engineered by subject matter experts. Acquire, QC, and transform centralized, annotated assay data, connected to clinical and sample data in a single connected ecosystem. With one click, bring intelligence to biomarker data with dynamic visualizations to surface trends across cohorts, patients, and timepoints to inform decision-making on-study. Explore data and generate insights through comprehensive analytics and reporting tools. Accessible Intelligence, Human-centric Design: eBDM, with this release featuring a reimagined user interface, enables cross-functional teams in data science, bioinformatics, translational research, and executive leadership to gain insights without data expertise. Combining Data Management with Business Intelligence: eBDM is a fully compliant (FAIR principles, 21CFR Part 11, GxP, GDPR) product, consisting of data management and business intelligence tooling. These tools enable quality data as the foundation for insight generation and information consumption through purpose-built analytics and reporting. eBDM is a knowledge amplifier for R&D. For example, teams can:
Navigate and explore cleaned, annotated data at scale across an entire portfolio—whether a single program or hundreds of studies, at all phases of development;
Generate multi-marker patient profiles to view attributes, e.g., tumor burden, over course of treatment to improve clinical trial design and decision-making;
This release features a new business intelligence engine with powerful visualizations and analysis options;
Improve repeatability by cross-referencing biomarker measures across file types; e.g., compare ctDNA profiles with efficacy biomarkers such as immunohistochemistry.
Revvity Signals | Signals Synergy (version 1.0) | Booth 519
https://revvitysignals.com/
Signals Synergy, a dedicated discovery informatics solution for pharmaceutical and biotech sponsors from Revvity Signals, replaces email, spreadsheets, and other inefficient methods of information sharing between sponsors and their CROs, CMOs, CDMOs, and academic labs. When added to Signals Notebook or Signals Research Suite, Signals Synergy overcomes challenges in communication, planning, and information exchange by eliminating the errors, wasted resources, and delays created by traditional systems.
Product launches in April 2024 at Bio-IT World and key capabilities include:
Administration: Simplified project initiation, user onboarding, and security setup
Drug Design: Ideation workspace for capturing drug designs and hypotheses
Intellectual Property Protection: Built-in masking of proprietary codes, properties, and material IDs
Project Management: Scientifically minded tools for tracking collaboration progress
Data Exchange: Automated transformation of unstructured CRO, CMO, and CDMO reports into structured data ready for analytics and visualization
Sapio Sciences | Sapio ELaiN - Electronic Laboratory Artificially Intelligent Notebook | Booth 817
https://www.sapiosciences.com/products/artificial-intelligence/
Sapio leverages large language models (LLM) that enable scientists to rapidly create experiments, generate code, ask for support, and search and visualize their data. In turn, scientists can achieve new levels of efficiency, speed, and insight—all with a simple chat interface that is built into their core lab informatics solution.
SciBite | SciBite Search 2.4, release is SciBite Chat | Booth 505
https://scibite.com/scibite-chat
SciBite Chat is a state-of-the-art search experience that gives you access to the most relevant information quickly and efficiently, allowing users to “have a conversation with their data.” The capability uses advanced ontology-based IR systems coupled with GAI to allow users to identify the answers they need in an explainable and reproducible way, without wasting valuable time sifting through irrelevant information. With a user-friendly interface that supports natural language queries, and an iterative chat experience, enabling true data democratization. Trust and traceability are at the core of the user experience; verbatim evidence highlighting is provided, as is the underlying query used to identify relevant documents; allowing users to easily pivot to the equivalent conventional search, supporting reproducibility as well as explainability. As SciBite Chat sits atop SciBite Search, users can be confident the underlying knowledge base is kept up to date and benefits from the document-level security, ensuring users can only chat to documents they are entitled to. SciBite Chat is fully customizable and can be tailored to your specific needs. Whether you prefer a SaaS solution or want to deploy it on-premises using your own OpenAI credentials, SciBite Chat gives you the required flexibility and control. With SciBite professional services, custom vocabulary curation, support, and advice, you can rest assured that you're getting the best possible support from a team of experts. Furthermore, like everything else at SciBite, the SciBite Chat interface is founded on a flexible API which can be directly integrated into existing systems.
Syntekabio | STB LaunchPad | Booth 22
https://cloud.syntekabio.com/
In November 2023, Syntekabio launched its AI Bio Supercomputer Center hosting 10,000 high-performance servers. Owned solely by Syntekabio, this private cloud infrastructure enables concurrently developing disease-agnostic drug pipelines for many protein targets. Supported by a CRO network, Syntekabio experimentally validates its in-silico predicted drug candidates providing a full cycle solution for drug discovery and development. This synergy notably shortens the discovery and development timeline reducing it to approximately two years to reach the IND-enabling stage for a drug pipeline. Syntekabio leverages its disease-agnostic AI drug discovery platform to continuously develop drug candidates. Its accelerated drug discovery program, STB LaunchPad, currently covers over 100 known target proteins spanning 60-70% of major indications. Through its strategic STB LaunchPad program, Syntekabio is providing a repository of pre-made hit and lead compounds available for off-the-shelf purchase. Protein targets in STB LaunchPad are proactively updated to include additional protein targets to meet dynamic market demands.
TetraScience | Tetra Data Platform v4.0 | Booth 304
https://www.tetrascience.com/
Brief Product Description & Technical Specifications:
The new 4.0 version of the Tetra Data Platform provides:
A foundation for Lakehouse cloud data storage that incorporates the best features of data warehouse and data lake architectures
Transformation of raw data into an open, common, vendor-agnostic data format
The largest and fastest growing library of purpose-built data integrations and data models
Data files enriched with scientifically relevant metadata using harmonized taxonomies and ontologies - Advanced search capabilities with role-based access controls
Low-code Python pipeline configurations for easy data movement and transformation
Reactive data automation that can be triggered by activities and events
Data version and lineage traceability through file journey audit trail
Co-location of data and your applications in a single workspace for the highest scientific productivity
AI and analytics readiness of previously inaccessible scientific data
The Hyve | Fairspace and 0.7.31-SNAPSHOT | Booth 301
https://www.thehyve.nl/services/fairspace
Fairspace is an open-source research data management platform that adheres to the FAIR principles. It provides a collaborative environment for managing any type of data and doubles as a data catalog and metadata repository. The platform employs a semantic metadata model in RDF, customizable by an organization or by The Hyve, to enable seamless semantic data integration. The Hyve is also able to build ETL processes to fetch metadata from various public or private data sources and repositories which researchers have zero time to do. Coupled with JupyterHub, it empowers you to analyze data and metadata using R, Python, or Julia without ever leaving the platform. The latest release From Fairspace boasts two new features:
A multi-modal domain with a single interface to search metadata from any type of biomedical domain anywhere from genomics to clinical trial data, images etc., despite their varied terminologies and data standards on the homepage. This allows researchers to go back and forth between the different domains. (readily available in the new version)
- A conversational retrieval augmentation mechanism, which processes queries in natural language, offering tailored summaries and returns relevant articles, maintaining a history of searches and adapting responses based on context. This feature aids researchers in categorizing studies or extracting specific information from extensive texts. This will be customizable to an organization especially when dealing with large amounts of free text data. The Hyve can offer consulting services to support this customization.
Velsera, Inc. | Pangenotyper | Booth 704
https://velsera.com/
Velsera and Sentieon introduce Pangenotyper, a groundbreaking NGS secondary analysis product that overcomes the limitations of traditional linear reference-based workflows. It couples Velsera's advanced pangenome aware read mapper with Sentieon’s DNAscope variant caller. Pangenotyper mitigates reference bias, improves complex variant calls, and reduces overall running cost compared to traditional NGS alignment and variant calling pipelines. This high-performance DNA analysis solution provides accelerated alignment speeds and proven, validated improvements in accuracy for detecting SNVs, indels, and structural variants from short read data. Pangenotyper is available through Velsera’s Seven Bridges platform as an optimized cloud application with an intuitive web-based interface, or through a stand-alone Docker image containing the workflow as a command line executable for users who prefer deployment in their private cluster or cloud. Velsera and Sentieon built a single, user-friendly experience that can readily connect to existing high-throughput workflows, and support common file formats. The Pangenotyper solution executes on standard hardware, and takes in single or paired end reads (presented as FASTQ, BAM or CRAM files) and a pangenome reference (GRCh38 or GRCh37). The workflow obviates the need for computationally expensive workloads, generates genotypes as fully compliant VCF files, and provides read mappings in standard BAM format. Pangenotyper also comes with genome browsing and graphical overviews of called genotypes for visual QC. The novelty of the product comes from the integration of a pangenome-aware aligner with an ML based variant caller, trained on the outputs of the graph reference. The performance of the combination is unprecedented.
Xybion | CompliancePredictor | Booth 224
https://www.xybion.com/quality-management-system/compliance-risk-predictor/
Xybion CompliancePredictor helps identify compliance risks using AI/ML capabilities. Here's an example report. (diagram) In this report it shows that Document control function has very low risk score and good (0.18, green) in the “process” area, but they are at a high-risk (0.95, red) in the “quality” area. In the software, the end user can double click on the specific area to find out what factors are driving the color coding. This will help the user to work on those specific areas to make everything good (green) before any internal or external regulatory audits. On double click, the user can also see if any of the issues have direct correlations with FDA 483s or linkage to any fines and settlement. This information comes from the Xybion database. These areas take very high priority for the users to close out before any FDA audits or DoJ actions. How does the system work? Xybion CompliancePredictor uses its own database of FDA issues, database of DOJ; and that can be augmented with company internal audit findings to create a benchmark. CompliancePredictor uses our patented (US patent#16/286,285 https://uspto.report/patent/grant/11,062,327) algorithm to predict risks - It has machine learning capabilities to constantly improve risk prediction confidence level. CompliancePredictor AI can read and monitor content on its own and finds out potential problems even before an audit happens. This helps the quality & compliance people proactively know about potential non-conformances and take corrective actions early. Data shows that an early action to avoid non-conformance can reduce the compliance cost by 1200%. CompliancePredictor can help the content creators as well to ensure they are not producing information or unknowingly creating non-conformances. This will avoid rework. This will also reduce audit time. This will reduce FDA findings. Here's another use case: CompliancePredictor automated document review capability (diagram) CompliancePredictor can read any document to find our compliance issues within seconds. Use cases are quality policies, SOPs, manufacturing records, study protocols, Medical/Legal/Regulatory review (marketing content), contact documents, SOWs, etc. Here’s an example of SOW read against predefined checklist. Document is placed in a sharepoint folder for RiskPrdictor to read and email the summary result to the end user within minutes. Manual review of this SOW may take 30 minutes, but system can do it within a minute. The automated document review capabilities have a very broad-based applicability to cover the entire pharmaceutical value chain from R&D to manufacturing to sales & marketing, both internally and to supplier network. Vendor audit documents can be reviewed through Xybion system to create a red/yellow/green heatmap that can help the auditors to focus on the right areas and get the audit done more efficiently. Unlike other automated reader software, Xybion can read, interprete and predict potential FDA issues coming from a document.