APECx Teaming Profiles

Thank you for your interest in ARPA-H’s Antigens Predicted for Broad Viral Efficacy through Computational Experimentation (APECx) program. This page is designed to help facilitate connections between prospective performer teams. If either you or your organization are interested in teaming, please submit your information via the form below. Your details will then be added to the list below, which is publicly available.

APECx anticipates that teaming will be necessary to achieve the goals of the program. Prospective performers are encouraged (but not required) to form teams with varied technical expertise to submit a proposal to the APECx. For questions, please contact us at APECx@arpa-h.gov.

APECx Teaming Profile Form

Please note that by publishing the teaming profiles list, ARPA-H is not endorsing, sponsoring, or otherwise evaluating the qualifications of the individuals or organizations included here. Submissions to the teaming profiles list are reviewed and updated periodically.

Interested in learning more about the APECx program?

Teaming Profiles List

To narrow the results in the Teaming Profiles List, please use the input below to filter results based on your search term. The list will filter as you type.

Organization Contact Information Location Describe your organization's current research focus areas Tell us what your organization can add to APECx and potential teaming partners Tell us about your organization's strengths and experience Tell us what your organization is looking for in potential teaming partners Which technical areas within APECx does your organization have the capacity to address?
JURA Bio, Inc. Elizabeth Wood (ew@jurabio.com)
Additional: es@jurabio.com
Worcester, MA At JURA, we have used machine learning to optimize the gene synthesis process, allowing us to create high-fidelity (ie, non-random) libraries of genes at previously unimaginable scales - 10^14+. We have used this extraordinary capacity to design and build sequences to probe antigen-TCR pairing at tremendous scale. We offer high-throughput protein engineering, as well as a protein modeling toolkit development for next-generation antigen design. We would hope to find teams gifted both in HT biochemical assays to use our ultra-large gene libraries as well as downstream clinical development partners. JURA was founded on principles of integrated probabilistic machine learning and synthetic biology with an emphasis on deployable, manufacturable technologies for advancing our understanding of immune repertoires. We have highly skilled immunologists, synthetic biologists, and machine learning and high performance computing experts. We would like to find teaming partners looking for a skilled preclinical integrated team combining best in class HT gene synthesis assays with best in class probabilistic antigen modeling to help us move our technology into translational candidate development and clinical evaluation.
  • Technical area 2: Protein modeling toolkit development for antigen design
Vaxess technologies, inc Livio valenti (Livio@vaxess.com)
Additional: Michael@vaxess.com
Cambridge, MA Vaxess is a clinical stage biotechnology company and leading developer of vaccine patches for skin delivery. Delivering vaccines via a patch it’s an attentive alternative to increase vaccine access but also to exploit the innate immunological strengths of the skin. We are the leading player in this space and we would lend our expertise in the translation of vaccine candidates into the patch format. We are the leading clinical stage vaccine patch developer, and have partnerships with large pharma companies such as AstraZeneca and others. We received previously darpa and barda funding and we have a gmp manufacturing line capable of producing phase 1 and 2 material for clinical trial in Boston. We would be interested in partnering with antigen developers that are using advanced computational models to advance novel antigen designs to be delivered in the skin with our patch.
  • Technical area 3: Translational candidate development and clinical evaluation
University of California Irvine Lisa Wagar (lwagar@hs.uci.edu)
Additional: ssureshc@uci.edu
Our team uses human immune organoid models to predict adaptive immune responses to vaccine and virus antigens. The organoids are generated from primary lymphoid tissues from diverse human populations and retain normal human immune heterogeneity. Organoids can be generated in high throughput and are amenable to many standard and novel readouts of adaptive immunity, including scRNAseq, repertoire analysis, flow cytometry, imaging, ELISA, high throughput protein array, and functional assays. We are building predictive and correlative models from >100 donors and numerous vaccine formats with a systems immunology approach. Using these models, our goal is to identify novel correlates of protection, identify host and antigen features that predict a robust vaccine response, and accelerate rational vaccine design efforts. It is our hope that human immune organoid approaches will help to bridge the current gap between pre-clinical models and human clinical trials. Although preclinical models (e.g. small animals and cell lines) are consistent and robust, they often fail to translate to efficacy in humans. We think a major source of this problem is that there are many points of heterogeneity in humans (age, sex, prior infections and vaccinations, chronic diseases) that are not adequately modeled with current preclinical studies. To address this, we can test numerous antigen and adjuvant combinations with reasonable scale (<10-50 per donor) in a model that retains human immune heterogeneity and diversity (100+ donors with diverse immune history and demographic factors). Our contribution to a potential team will be deep knowledge of the immune microenvironment of human immune cells and lymphoid tissues, and expertise in big data analysis and modeling approaches. The Wagar lab resides at the University of California Irvine (UCI). UCI is uniquely situated for multidisciplinary and collaborative efforts in vaccine immunology, with abundant interactions between the Department of Physiology and Biophysics, Institute for Immunology, Center for Virus Research, and the Vaccine R&D Center. We are highly collaborative and excited about team science. Our combined knowledge of sophisticated wet-lab and dry-lab techniques would allow us to plug into multiple stages of a vaccine development and testing program. We have access to BSL2+ and BSL3 facilities at UCI, pending biosafety approvals. We are looking for teaming partners with new ideas for antigen design and delivery. Although we have numerous tools to assess the immunogenicity of novel candidates and make empirical measurements, we do not have expertise in antigen design. Much of our work is focused on understanding T cell and antibody breadth to influenza and we would be excited to work with a team that is also focused on development of a universal influenza vaccine. However, we are open to working on other viral threats with a knowledgeable team.
  • Technical area 1: High-throughput biochemical analysis and protein engineering
Valneva Pamela Duchars (pamela.duchars@valneva.com)
Additional: olivier.jankowitsch@valneva.com
Vienna, Austria Valneva is a specialty vaccine company focused on developing prophylactic vaccines against infectious diseases. Valneva has end-to-end capabilities to bring innovative vaccine candidates to licensure. Valneva brings the preclinical, technical, and clinical development expertise and experience to bring innovative vaccine candidates through all stages of development. Valneva has proven experience bringing vaccine candidates to licensure, including IXIARO (against Japanese encephalitis) and the first COVID-19 vaccine to receive full market authorization in Europe. The company has a vaccine against chikungunya in review with the FDA for licensure in the US and a vaccine against Lyme disease in phase 3 clinical efficacy trials (partnered with Pfizer).

The company has state-of-the-art expertise in preclinical, technical, and clinical development and the facilities to perform all preclinical development, process development, and GMP manufacturing in house.
Valneva is primarily interested in leverage our expertise in product development and work with partners who design innovative vaccine candidates in infectious disease areas with unmet medical need.
  • Technical area 3: Translational candidate development and clinical evaluation
1 - Theradaptive Luis Alvarez (luis@theradaptive.com)
Additional: david.stewart@theradaptive.com
Boston, MA and Frederick, MD Theradaptive has a protein engineering platform that permits rapid screening of protein variants in biochemical or cell-based formats. Theradaptive offers its engineered MHC Class I and Class II receptor / antigen binding platform to enable rapid screening of designed antigen candidates. Theradaptive has extensive expertise in the following:

1. protein engineering of immunologically important proteins
2. development synthetic MHC particles that present antigen to enable HTS
3. development of analytical biochemical and cell-based assays
4. GMP manufacturing
We seek to partner with teams that have expertise in the design of novel antigens, or rational or computational antigen design.
  • Technical area 1: High-throughput biochemical analysis and protein engineering
ProtaBody, Inc. Barry Olafson (barry.olafson@protabody.com)
Additional: justin.chartron@protabody.com
Pasadena, CA We are an antigen/antibody protein engineering that uses equal parts of computational protein design, ultra-high throughput protein expression and characterization, and AI and machine learning to optimize the activity, structure conformation, stability and aggregation properties for antigens and antibodies. The key is we design and experimentally assay large scale diverse protein sequence libraries (tens of millions) that are optimized to provide the needed wide-ranging positive and negative data for machine learning over vast regions (10^30) of sequence space. We can explore large combinations of simultaneous mutations in any or all of the CDRs and framework regions for full-length IgGs and even larger antigens to generate in-house the training data that AI and ML need to correlate function with protein sequence. With this unprecedented scale, we have repeatedly succeeded at difficult tasks, such as generating potent, broadly neutralizing antiviral antibodies by re-engineering paratopes to accommodate multiple variants of epitopes. Viral antigens are a moving target due to their high mutation rates, and effective antigen design should attempt to anticipate potential escape mutations. Single mutant scans provide a starting point for determining which escape mutations are available to the virus, but assessing combinations of potential single mutant escape mutations on the antigen will generate a more robust sequence profile, particularly when the human cell receptor is known. Our AI-based antigen engineering platform is capable of exploring the antigen sequence landscape to anticipate future escape variants. We can also provide additional antigen engineering collaborations on other aspects of antigen design, including expression levels, stability and reduced aggregation. We can also provide collaborative efforts for the modeling and simulation of conformational flexibility of putative antigens. Also, our high-throughput experimental platform can be used to validate various aspects of an antigen’s physical manifestation and activity. We have successfully engineered a number of COVID-19 parental antibodies 1) to increase potency, 2) broaden the range of neutralization to multiple SARS-CoV-2 variants, and also SARS-CoV-1 and 3) rescued antibodies that lost potency as the virus mutated to previous levels of neutralization for new variants. Our TRIAD Computational Protein Design (CPD) software platform has been developed entirely by us and continues to be enhanced as the need arises. Our ultra-high throughput experimental platform involving yeast display, FACS sorting experiments and long-read sequencing is all performed in-house in its entirety with design cycles involving millions of trial sequences generally completed in a matter of a few weeks. We are also developing novel engineering capabilities for bi-specific antibodies. We will need partners to incorporate our designed and engineered antigens into their vaccine platforms for validation and advancement to clinical trials. We are also looking for partners who can discover parental antibodies through immunization in mice or other animals that we can use in our experimental verification of potential escape variants. We are also willing to collaborate with companies who would benefit from using our TRIAD CPD software, which has been licensed to large pharmaceutical and biotech companies.
  • Technical area 1: High-throughput biochemical analysis and protein engineering
Draper Cassie Bryan (cbryan@draper.com)
Additional: John Julias, jjulias@draper.com
Cambridge, MA Draper is a not-for-profit research organization focusing on a wide range of technical and application areas, including space and navigation, electronic systems, biosurveillance, applied biology, and medical countermeasure (MCM) development. More specifically related to APECx, we have programs and expertise in protein engineering, organ-on-chip, tissue models of infectious disease, multi-omics analysis, AI/ML model development and application, and computational protein design and modeling. Draper has an automated, integrated bioinformatics pipeline for analysis of conformational epitopes that are conserved at the viral family level. This pipeline has been built with scalability and tunability in mind to enable analysis of the entire viral proteome and customizability for multiple applications. This pipeline consists of multiple steps, including data gathering and input preparation, epitope prediction, conservation analysis, clustering, and scoring. We also have a strategy to incorporate T cell epitope prediction as well as computational protein design to generate an optimal, stable antigen that supports these identified epitopes for development of a broad-spectrum vaccine. Draper has a cross-functional staff of engineers in a wide range of technical areas, including protein engineering, design, and modeling, as well as bioinformatics, microbiology, and immunology. We have infrastructure and capabilities to support high-performance cloud computing, AI/ML, and software development. Draper has extensive experience managing and executing milestone-driven contracts for a variety of government customers. Draper is looking for teaming partners with capabilities in high-throughput generation of large protein structure and function data sets as well as additional properties that are important to vaccine success, such as immunogenicity, manufacturability, and stability. Experimental data sets will be crucial in development and optimization of a complete bioinformatics pipeline that is able to predict and design all aspects of protein vaccines.
  • Technical area 2: Protein modeling toolkit development for antigen design
University of Virginia Steven Zeichner (zeichner@virginia.edu)
Additional: slz7b@uvahealth.org
Charlottesville, VA We are developing a new vaccine platform: Killed Whole Cell (KWC) Genome-Reduced Bacteria (GRB). We use an inducible Gram-negative autotransporters in a synthetic plasmid we designed to place vaccine antigens on bacterial surfaces, then inactivate bacteria to make KWC/GRB vaccines. GRB have a large fraction of their genomes deleted, which enhances antigen exposure. We published an initial description of the technology demonstrating efficacy in preventing clinical disease in an animal model: https://doi.org/10.1073/pnas.2025622118.

The platform offers significant advantages for pandemic/biothreat response:

1. With advances in synthetic biology, we can make a testable vaccine in ~ 3 weeks from target identification.

2. Factories that produce conventional KWC bacterial vaccines against diseases like pertussis and cholera already exist around the world, enabling rapid deployment.

3. Conventional KWC vaccines are very inexpensive (e.g. <$0.25/dose for pertussis).

4. Feedstocks for KWC bacterial vaccines are abundant and inexpensive.

5. KWC bacterial vaccines are stable: 24 months at 2-6 C.

6. The rapid, inexpensive characteristics of KWC/GRB vaccines enables classical engineering design/build/test/learn approaches to vaccine antigen optimization.

7. Some KWC vaccines are administered orally and elicit mucosal immunity.

8. Low cost, stability, ease of manufacture make vaccines produced using KWC/GRB ideal for One Health (human+animal) use.
Several groups involved in APECx are working to develop ways to identify new pathogens, including those with pandemic and biothreat potential. Other groups involved in APECx are involved in identifying tumor-specific antigen targets. Our KWC/GRB platform offers a clear pathway to rapidly make cost-effective and implementation-friendly vaccines using that knowledge. The University of Virginia is a large state university with excellent basic science, engineering, and clinical expertise and resources. Our group invented the KWC/GRB platform. We have unique experience and capabilities in using the platform and are employing the platform to develop several candidate vaccines, and are working to further advance the technology. Our new vaccine platform is ideally suited to respond rapidly to new pandemics (disease X) and emerging biothreats. We can produce a testable vaccine candidate in ~3 weeks from target identification. It is also well-suited to rapidly producing low cost custom cancer immunotherapy. We can rapidly produce new candidate vaccines, so we would like to partner with groups that can rapidly identify new diseases and likely targets in the pathogen for vaccine development. That antigen target information could then feed into our platform to enable the rapid production of a new vaccine.
  • Technical area 1: High-throughput biochemical analysis and protein engineering
Early Charm Ventures Nathan Nimbargi (nathan.nimbargi@earlycharm.com)
Additional: stephen.farias@earlycharm.com
Baltimore, MD Early Charm specializes in a diverse portfolio of drug design-focused companies, dedicated to supporting computational researchers in streamlining the drug design process. Our research encompasses various areas, including protein simulations, the mapping of protein-ligand interactions, precise identification of binding sites, free energy calculations, structure-based drug design, small molecule parameters, and accurate protein protonation state predictions. One of our notable research achievements is the development of innovative target molecule simulation software, employing GCMC(Grand Canonical Monte Carlo)/MD(Molecular Dynamics) simulations in combination with probe molecules and water solvents. This groundbreaking approach produces a pre-computed FragMap, providing valuable insights into protein characteristics and offering data on the likelihood of specific functional groups residing in a protein site. Another significant aspect of our research focuses on enhancing the accuracy of protonation state information for proteins. Additionally, our efforts extend to the mapping of GPCR proteins and the exploration of allosteric binding sites on proteins. At Early Charm, we are also equipped with the capability for high-throughput virtual screening, a valuable tool for screening compounds and preparing them for experimental testing. Our work on advancing drug design methodologies is at the core of our research efforts. We provide computational resources in the form of computational software, enabling researchers to fine-tune their drug design strategies well before entering the laboratory. Our comprehensive suite of tools includes:
a) Precise knowledge of optimized protein-ligand complexes through our SILCS-Monte Carlo Docking Software.
b) Identification of potential protein binding sites using SILCS-Hotspots.
c) Efficient compound screening via SILCS-Pharmacophore.
d) In-depth exploration of protein-protein interactions and biological formulations with SILCS-Biologics.
e) Parameterization of small molecules, utilizing the CHARMM force field, powered by our CGenFF program.
f) Accurate predictions of protonation states for proteins through iTitrate.
Our portfolio companies SilcsBio and ComputChem can aid companies in the drug design space.
Early Charm has extensive experience collaborating on a diverse range of projects, including government-funded grants and service initiatives. Our grants team actively seeks collaborative opportunities with other companies, offering the flexibility to serve as either the primary contractor or subcontractor as the need arises. Our in-house team comprises legal experts, grant writers, scientists, engineers, sales, and administrative staff, all working collaboratively to drive projects forward. A significant portion of our work involves the strategic commercialization of technology. We specialize in acquiring intellectual property at early technology readiness levels and transforming it into commercially viable products and services. Leveraging our team of computational chemists and adept developers, we advance technology in the drug design field and provide valuable services for various projects. Notably, Early Charm has successfully commercialized products such as SILCS, a drug design software for predicting binding energies, and iTitrate, a software that accurately predicts protonation states in proteins. We would like to team with partners looking for computational services in the drug design space. Our focus lies in running simulations on target molecules, fine-tuning proteins with diverse parameters, and providing comprehensive computational support to expedite the drug development process. In addition, we are actively seeking partners interested in joint ventures to harness grants for the commercialization of technology within the field of drug design.
  • Technical area 1: High-throughput biochemical analysis and protein engineering
Mechanologics Matthew Lang (matt.lang@vanderbilt.edu)
Additional: hwm@tam.edu
Nashville, TN During immune surveillance in vivo, force derived from cell motions and cell-cell interface embedded actin-myosin machinery enhances T-cell receptor (TCR) binding and specificity, permitting certain cytolytic T lymphocytes (CTL) to recognize scant copies of a ligand on a diseased cell. We are currently focused on the extraordinary mechanism underlying how abTCRs exploit dynamic mechanosensing to achieve force-mediated recognition of their ligands while ignoring self. We are particularly interested in biophysical principles driving selection of the very best T cells, which we term “digital T cells” and defining measurable and/or computational parameters predicting digital performance of a T-cell against a particular target molecule. Our collective team integrates a cycle of measure, make and model where TCR-pMHC pairings under load are studied through experimental, computational and functional assays to assess the biological impact at single molecule, single cell and organism levels. Measurement capabilities for probing TCR-pMHC pairs under load: our single molecule and single cell assays classify which TCRs of a T-cell repertoire directed toward an antigen manifest digital mechanobiological performance. Teaming partners will have candidate TCR-pMHC pairs that perform digitally. Our goal is to not only identify digital pairings but identify biomarkers that correlate with biophysical TCR performance parameters. Modeling capabilities for evaluating the TCR-pMHC bond under load: Molecular dynamics under force (MDf) will analyze TCR-pMHC repertoires to determine whether interfacial binding maintains organization and strengthens or releases under physiological pN load. Teaming partners will have structural data against common peptide to train models that predict which TCRs behave digitally. Use of Poisson detection liquid chromatography data independent analysis mass spectrometry methods: This will be used to identify antigens actually being presented and quantify their level of luminosity or sparsity. Teaming will direct which targets to focus on. We are particularly strong in mechanobiology and physical insight towards early T cell activation. We were the first to define the TCR as a mechanosensor, offering a physical solution to the longstanding question of how T cells can achieve rapid and specific sensing of a single peptide bound to an MHC molecule among a sea of unrelated peptides arrayed on an antigen presenting cell surface. Our team has been collaborating with each other for years and have optimized assays and methods in place for identifying and characterizing digital cells. Dr. Reinherz at Dana-Farber Cancer Institute has revealed key functional and structural discoveries about T cell receptors including CD3 signaling subunit components and how TCRs bind to peptide-loaded MHC in conjunction CD4 and CD8 co-receptor molecules. Dr. Lang at Vanderbilt, a single molecule biophysicist, with expertise in optical tweezers and single molecule technologies has developed assays to elucidate the structural transition and biophysical parameters of TCR-pMHC bond formation leading to early T cell activation. Dr. Hwang at Texas A&M, a computational biophysicist and CHARMM developer, has elucidated the atomistic basis of TCR-pMHC catch bond formation using molecular dynamics simulations that can probe the effects of physiological level pN load. We are interested in studying and advancing the overall strategy of identifying digital cells and eliciting these through mechanobiology guided vaccine development. Please see Reinherz, Hwang and Lang PNAS 2023 “Harnessing ab T cell receptor mechanobiology to achieve the promise of immuno-oncology.” for a recent perspective. In Technical area 1, we are interested in assaying digital TCR-pMHC pairs. In Technical area 2, we are interested in advancing methods for modeling the TCR-pMHC structure under loaded conditions. In Technical area 3, we are interested in evaluating TCR-pMHC pairs that are protective and guiding the performance of vaccine strategies or cancer immunotherapy in elucidating T cell repertoires containing protective digital T cell performers.
  • Technical area 1: High-throughput biochemical analysis and protein engineering
Netrias, LLC Mohammed Eslami (meslami@netrias.com)
Additional: weston@netrias.com
Annapolis, MD and Cambridge, MA We focus on the development of automated data harmonization pipelines of heterogeneous, multimodal data sources from public and private repositories. Our technology platform uses GenAI to harmonize data and metadata from experiments into well, curated data repository. The technology was built on a DARPA STTR and has transitioned to CDC, multiple institutes at NIH, and DTRA. We can lead the development of harmonization pipelines of data collected on the program as well as those available in public repositories into a well-curated database. Our technology is modular and adaptable so that it can serve as a tool that can be used well after the current effort is also complete. The use of GenAI will reduce the burden on TA2 performers to search for data they want to use generated by TA1 groups as well as those available in public repositories. Our proven automated technology gets life sciences data AI ready lightning fast with minimal human curation to advance the scientific discovery process. We allow scientists to focus on the science and analysts to focus on the models and we take care of everything in between. Data is actually FAIR when it passes through our platform and can be seamlessly queried and integrated with other relevant data. We are interested in teaming with people looking to pursue a TA2 ONLY bid. We can support their modeling efforts by building an automated curation process across the TA1 performers that will be generating data. We can also support making the models and other software tools they build accessible to a wider community.
  • Technical area 2: Protein modeling toolkit development for antigen design
Michigan State University Guowei Wei (weig@msu.edu)
Additional: weig@msu.edu
East lansing We are working on a few integrated multidisciplinary research areas:
1) AI-based forecasting of emerging pandemics.
2) AI- and math-assisted antigen design for emerging pandemics.
3) AI- and math-assistant protein engineering and directed evolution.
4) From single molecule to single call, multi-omics, multiscale, multiphysical, multidisciplinary approaches to emerging infections and potential pandemics.
5) The development of algebraic topology, geometric algebra, differential geometry, and combinatorics-based novel tools to revolutionize computational biology and AI.
We have unique expertise in computational biology. My team integrates cutting-edge artificial intelligence (AI) techniques, advanced mathematics (i.e., algebraic topology, differential geometry (DG), and combinatorics), genotyping of viral genomes, a variety of molecular biophysical techniques (QM/MM, MD, docking, homology modeling, multiscale modeling), bioinformatics, proteomics, genomics, immunomics, and quantitative systems immunology/pharmacology. My team was the top winner in the D3R grand Challenges, an annual worldwide competition series in computer-aided drug discovery, funded by NIH (https://users.math.msu.edu/users/weig/D3RGC2.pdf; https://users.math.msu.edu/users/weig/Wei_Team.pdf; https://users.math.msu.edu/users/weig/D3R_GC4.pdf). My team was invited to form partnerships with Pfizer, Bristol Myers Squibb, and FDA (MOU 225-23-007 | FDA) on drug discovery. My team has a proven record on the discovery of SARS-CoV-2 evolutionary mechanisms and successful forecasting emerging dominant viral variants about two months in advance. 1) We predicted two mutation sites 452 and 501 on the viral spike in May 2020 (https://arxiv.org/pdf/2005.14669.pdf ), which had been proven to host the key mutations of all prevailing variants, alpha, beta, gamma, delta, theta, mu, and omicron, BA.2, BA.4, and BA.5, etc.

2) We discovered SARS-CoV-2 evolution mechanism, natural selection via mutation-induced infectivity strengthening in summer 2020 (https://doi.org/10.1016/j.jmb.2020.07.009). To my best knowledge, my team was the first to report this mechanism with a spike-ACE2 model, AI prediction of binding affinity changes upon mutation, and genotyping of viral genomes.

3) We discovered another SARS-CoV-2 evolution mechanism, natural selection via antibody resistance (or vaccine breakthrough) in late 2021 (https://doi.org/10.1021/acs.jpclett.1c03380?urlappend=%3Fref%3DPDF&jav=VoR&rel=cite-as). To my best knowledge, my team was the first one to report this mechanism with the AI predicted mutational disruptions of 135 antibody-spike binding complexes and the correlation between virus mutations and vaccination rate in many countries.

4) Based on two mechanisms, we successfully forecast the emerging dominance of Omicron BA.2 in the early February of 2022 (https://arxiv.org/abs/2202.05031), that was confirmed by the WHO in late March 2022.

5) We also forecasted the emerging dominance of Omicron BA.4 and BA.5 on May 1st, 2022 (https://arxiv.org/pdf/2205.00532.pdf), that was confirmed by the WHO in early July 2022.
We are looking for experimental partners in the design and test of mRNA vaccines, the deep mutational scanning of antibody-antigen binding complexes, in vivo/in vitro experimental measurements of mutation-induced antibody-antigen binding free energy changes. Our goal is to develop effective preventive vaccines for the emerging pandemics in the future.
  • Technical area 2: Protein modeling toolkit development for antigen design
Macrotope, Inc. Stephen Anderson (steve@macrotope.com)
Additional: elliot@macrotope.com
Princeton, NJ We have decades of experience in protein engineering, high-throughput protein expression, and biochemistry, and leverage this expertise to solve fundamental problems in antigen design and therapeutic antibody discovery. Our current research focuses on improved vaccines for endemic and emerging diseases, as well as therapeutic antibody discovery for infectious disease and oncology. We take an antigen-centric approach to immunology, and have repeatedly demonstrated that careful antigen selection and design greatly improves the probability of success in manufacturing, scale-up, immunogenicity, and efficacy. We offer proven and unparalleled experience with antigen construct design and high-throughput antigen production, and have pioneered next-generation antigen engineering technologies that elicit broad and potent immune responses and increased antibody diversity. Our Antigen Design for Manufacturing approach produces antigens that are both optimized for manufacturability and scale-up, and are highly immunogenic in order to activate a robust adaptive immune response. We have extensive experience and demonstrated results in all three Technical Areas identified by APECx. Macrotope’s founders have extensive experience in high-throughput protein production both for structural determination (as part of the NorthEast Structural Genomics Consortium), and antigen production (as part of the NIH Protein Capture Reagents Program). The latter effort involved designing, expressing, purifying, formulating, and shipping over 1000 unique human transcription factor antigens to multiple partners for antibody generation. Through this effort, we developed a highly optimized antigen construct design workflow that was continuously refined based upon successful antigen production and antibody generation. Our antigen construct design algorithm works even in the absence of structural information for the target antigens, and is well suited to create stable, highly immunogenic constructs targeting conserved, potentially broadly neutralizing regions in target pathogen families. We are looking for partners with experience translating pre-clinical stage products to the clinic, including cGMP vaccine production, antibody development and lead optimization, clinical trial administration, and IND filings. We are also looking for collaboration partners who can benefit from our HTP protein production and antigen design expertise or proprietary antibody discovery platform.
  • Technical area 1: High-throughput biochemical analysis and protein engineering
Mayo Clinic Vaccine Research Group Rick Kennedy, PhD (Kennedy.rick@mayo.edu)
Additional: Feeder.Scott@mayo.edu
Rochester, MN The Mayo Clinic Vaccine Research Group uses mass spectrometry approaches to identify viral pathogen-derived and naturally presented peptides. We then test these peptides for immune recognition in vaccinated/convalescent human samples. This process allows us to ‘see what a T cell sees’ and identify immunologically relevant epitopes for peptide and protein-subunit vaccine design and development. We have successfully done this with influenza, measles, mumps, smallpox, zika, and SARS-CoV-2. We also have extensive experience using systems vaccinology tools (transcriptomics, epigenomics, single cell technologies, proteomics, flow cytometry) and approaches to characterize, understand, and predict human immune responses (both humoral and cellular) to viruses and vaccines. We offer: 1) a highly successful human vaccine research laboratory with a 4 decades-long history of NIH funding with 700+ publications in the vaccines and public health space, 2) significant and long-standing expertise in human immunology, 3) long-standing expertise and experience in preclinical (and animal) development and laboratory evaluation of viral vaccines, and 4) the extensive laboratory and clinical resources of the Mayo Clinic for designing and executing clinical trials for evaluating vaccine candidates. Strengths: Access to world-class clinical facilities, infrastructure, Core laboratory resources. An expert team with a broad array of humoral and cellular immunology experience combined with expertise in various ‘omics’ technologies. Established ties to populations that recognize the value of research and readily engage in clinical trials.

Weakness: We do not have expertise in protein or carbohydrate antigen engineering, nor in the development of new adjuvants.
We are looking to create a collaborative, multi-disciplinary team with cutting-edge expertise using tools of the new biology, computational modeling, and AI. To do this, we seek partners with expertise in technical areas 1 and/or 2. For example: a team with expertise in peptide antigen design that is looking to complete preclinical testing and move promising candidates into clinical trials. Alternatively, a team with bioinformatics and computational modeling (particularly AI) expertise that can materially enhance our mass-spectrometry based approach and create, enhance, or refine broadly reactive viral epitopes capable of eliciting protective immune responses to multiple pathogens within (or across) viral families.
  • Technical area 3: Translational candidate development and clinical evaluation
Teledyne Scientific & Imaging Aaron Mahler (aaron.mahler@teledyne.com)
Additional: michael.weisend@teledyne.com
Durham, NC We currently focus on neuroscience AI/ML research. In neuroscience we have experience in decoding complex neurophysiological signals for decision making and neuromodulation. Our AI/ML capabilities include bleeding-edge research in diffusion maps and graph-based approaches for extending the current capabilities and mitigating the limitations of neural nets. We have also recently explored ways to improve chemical detection in collaboration with FLIR. Our expertise in graph based modeling provides us with the tools to iterate on existing graph-based molecular modeling techniques. We aim to leverage our in-house experts in quantum and analytical chemistry in conjunction with our AI/ML expertise to create novel solutions for molecular modeling and the properties these models can accurately predict. Teledyne has extensive experience in performing on DARPA programs related to AI/ML and neuroscience. Our fabrication facilities as well as local compute cluster resources facilitate the rapid advancement of novel research ideas. We have personnel with backgrounds in quantum chemistry and analytical chemistry relevant to molecular modeling and property prediction. We are looking for partners seeking to propose solutions to TA1 and TA3 of APECx.
  • Technical area 2: Protein modeling toolkit development for antigen design
EpiVax Inc. William Martin (martinb@epivax.com)
Additional: annied@epivax.com
Providence, RI Immunogenicity assessment of vaccine antigens and therapeutic proteins
Development of T cell-based vaccines against infectious disease and cancer.
Development of T cell-based therapies related to allergy, auto-immune disease, and cancer.
Class I and Class II T cell epitope discovery, classification (effector vs regulatory), and antigen design for vaccines targeting infectious disease and cancer.
Ex-Vivo testing for HLA binding and T cell activation.
Evaluation of immunogenicity of antigens and candidate vaccines in HLA-Transgenic mice.
25 years of commercial success in the fields of epitope discovery, immunogenicity assessment, and vaccine design.
Strong track record with respect to grant development and funding.
EpiVax would like to partner with a larger (lead) organization capable of coordinating activities and addressing technical area 3.
  • Technical area 2: Protein modeling toolkit development for antigen design
GiwoTech, Inc. Ashwin Lokapally (ashwin@giwotech.com)
Additional: razvan@giwotech.com
Boston/Cambridge, MA GiwoTech is a TechBio platform company accelerating therapeutics and vaccine discovery by building digital twins of viruses. Our physics (molecular dynamics - MD), artificial intelligence (AI)/machine learning (ML), and enhanced sampling based simulations can provide dynamic protein-protein interactions, multi-protein-ligand interactions, identification of hidden and allosteric binding sites, structure-based drug design, free energy, and IC50 calculations. We have expertise in computational biology, bioinformatics, artificial intelligence (AI)/machine learning (ML) techniques, along with advanced mathematical and biophysical techniques
1) We can provide modeling of viral proteins/antigens and characterization of native and mutant viral proteins.
2) 3D structural information on viral proteins or protein complexes to help develop therapeutics, vaccines and neutralizing antibodies (nAbs).
3) Simulation of the whole virus particle to understand viral dynamics in interaction with small molecules.
4) Efficient screening, prediction and parameterization of small molecules and nAbs.
GiwoTech's consists of a highly experienced interdisciplinary team with skills in molecular biology, biochemistry, chemistry, biophysics, computational biology and AI/ML. We have capabilities to support high-performance computing and access to one of the best high performance computing clusters. We are seeking partners from Technical Areas 1 & 3 interested in collaborations towards commercialization.
a) Teams having expertise in HT functional analysis, biochemical data, immunological assays and protein engineering.
b) Teams capable of doing translational candidate development through lead optimization and validation, functional assays in animal models or equivalent systems to determine immunogenicity, safety, and efficacy, as well as developability and manufacturability.
  • Technical area 2: Protein modeling toolkit development for antigen design
SandboxAQ Lloyd Dabbs (lloyd.dabbs@sandboxaq.com)
Additional: andrea.bortolato@sandboxaq.com
Tarrytown, NY Drug discovery remains a long, costly process that leads to failure more than 80% of the time. SandboxAQ is at the forefront of transforming drug discovery using a combination of molecular simulation and AI. We harness the unmatched capabilities of GPU-driven AI and emerging quantum technologies to address the key bottlenecks and failure points in drug discovery and development.

SandboxAQ’s approach focuses on:

1. Molecular Optimization, small molecules, and biologics: Through the use of data generated via physics-based and quantum simulations, SandboxAQ’s machine learning models provide invaluable insights, particularly when confronted with novel targets with limited experimental data. This empowers SandboxAQ to identify potential breakthroughs even for historically “undruggable” targets.

2. Critical Application: SandboxAQ’s groundbreaking method involves ML training with data from molecular dynamics-based absolute free energy perturbation methods to predict the binding affinity between molecules. This enhanced technique delivers unparalleled accuracy and speed for hit finding and optimization.

3. Innovative Custom Software Solutions: we build custom software for different stages of drug discovery and development, bringing together physics-based methods and the latest in ML. Our current tooling focuses on molecule design, including structure-based antigen evaluation, based on absolute free energy perturbation methods.
SandboxAQ’s strategy is to progressively connect pharmaceutical R&D to the exponential scaling of GPUs via emerging novel technologies. Specifically, SandboxAQ works on molecular optimization, target validation, and clinical development.

Despite the progress in protein structure predictions, like AlphaFold, these methods are limited in their ability to predict the effect of mutations, protein-protein complex conformations, and binding affinities. SandboxAQ’s solution enhances these technologies using innovative physics-based all-atom simulations. This solution based on molecular dynamics refines the initial 3D models to score their stability and protein-protein free energy of binding accurately and quickly.

SandboxAQ’s active-learning-enhanced AQ-FEP solution is a unique absolute free energy perturbation workflow that efficiently explores the chemical space to identify ligands with optimal binding affinity to the target system by screening thousands of protein mutations. It also unlocks the possibility of training an ML model to efficiently expand the search to billions of protein sequences. This approach maximizes the probability of success in finding potential game-changing antigens, minimizing the risk of false positives and false negatives.

SandboxAQ’s solution can be valuable for antibody optimization to maximize the binding affinity to the selected antigen. Similarly, it could be applied across multiple relevant dimensions such as specificity, and absence of immunogenicity.
We have had significant success in supporting drug discovery efforts in neurodegenerative diseases and oncology.

In particular, SandboxAQ’s PrionDock platform, developed for the laboratory of Stanley Prusiner (a Nobel laureate), is a custom solution optimized to predict the functional activity of small molecules acting with an unprecedented innovative mode of action. The custom solution was able to result in unprecedented impact where other state-of-the-art computational tools failed. PrionDock unlocked for the first time rational structure-based drug discovery for this uniquely challenging class of proteins. Virtual screening based on this innovative technology resulted in an unprecedented hit rate, over 10 times superior to prior high-throughput screening campaigns.

SandboxAQ published a research article titled “Tensor Processing Units as Quantum Chemistry Supercomputers”, available at https://pubs.acs.org/doi/10.1021/acs.jctc.2c00876

SandboxAQ's scientific advisory board comprises:

- Dr. Geoff Ling - CEO, OnDemand Pharmaceuticals, Professor of Neurology, Johns Hopkins, DARPA Biotech

- Dr. Samir N Khleif - Immunologist, Oncologist, Drug Discovery, NIH, Georgetown, Georgiamune

- Dr. Siddhartha Mukherjee - Medical Oncology, Columbia

- Dr. Mark Smith - Head of Medicinal Chemistry, Stanford, Sarafan ChEM-H, Roche

- Dr. Steven Deitcher - CEO Bespoke Biotherapeutics

- Dr. Patricia Weber - SBDD expert
SandboxAQ is seeking partners in technical areas 1 and 3.
  • Technical area 2: Protein modeling toolkit development for antigen design
GE HealthCare Technology and Innovation Center John Nelson (john.nelson@gehealthcare.com)
Additional: guay@gehealthcare.com
Niskayuna, NY Our team is developing automated DNA-based and RNA-based expression construct production capabilities that dramatically reduce time and effort. Our team has experience in developing molecular biology, chemistry, mechanical engineering, and automation solutions for simplifying complicated laboratory workflows. Our team has an extensive background in development of workflows that utilize nucleic acids. We are interested in participating as a subcontractor, assisting with rapid production of DNA or RNA constructs in a manual or automated workflow.
  • Technical area 1: High-throughput biochemical analysis and protein engineering
Serimmune John Shon (john.shon@serimmune.com) Goleta, CA Serimmune has a technology, Serum Epitope Repertoire Analysis (SERA), that uses a 10 billion member random peptide bacterial display library that, coupled with Next Generation Sequencing and machine learning methods, enables a robust characterization of the entire B cell epitope repertoire in individuals and populations. We have used SERA to elucidate epitopes elicited by a wide variety of conditions and have developed over 50 research diagnostic panels in infectious diseases, autoimmune disease, and cancer. Furthermore, we are able to identify amino-acid residue level binding interactions as well as quantitatively track changes in epitope signals over time. This capability has been used to characterize epitopes elicited by a variety of vaccines, including those using mRNA, adenovirus, ferritin nanoparticles, and live attenuated platforms, in pre-clinical models and clinical trials. Finally, we have applied the same assay to over 20,000 individual subjects to create a population database with healthy and disease states from a variety of geographies - the database enables robust normalization and power to detect highly specific outlier antigen/epitope signals. In summary, we enable highly multiplex, high resolution and quantitative characterization of antigens and epitopes with computational methods that improve with every sample added to our resource. Serimmune’s capabilities will benefit partners in providing both empirical data as well as in silico models to understand the immunogenicity of antigens in both natural infection as well as vaccines. 1) For any virus genus of interest, we can either mine our database or process serum samples from disease cohorts to identify and prioritize the most significant antigens and epitopes associated with disease in an unbiased, data-driven way. Pre-existing signal in subjects and populations to understand existing/ imprinting epitope signals can also be performed. Both linear and non-linear epitopes are characterized by our system, and we are developing methods to map our non-linear epitopes to structures through experimental data and collaborations with partners. 2) For any vaccine formulation given to animals or humans, for any condition, we can broadly assess B-cell epitopes elicited by the vaccine. 3) The random peptide library A) enables us to obtain the specific composition of residue positions with logos plots - identifying conserved residues that are critical for binding B) in combination with mutation databases enables the identification of variants that potentially result in loss of immunogenicity, and C) enables the development of general immunogenicity models that might be applied to antigen design. Serimmune has a dedicated and highly experienced staff that has decades of experience working on epitope identification and resolution. We have expertise from scientists who initially developed the technology at UC Santa Barbara as well as computational scientists who have decades of experience in bioinformatics, translational informatics, machine learning and AI. The team have worked with multiple multinational pharmaceutical companies as well as leading academic and government institutions and have recently published many articles in journals such as Nature, Nature Medicine, Science Immunology, Journal of Infectious diseases, Frontiers in Immunology, etc. We have developed panels in a wide variety of viral infections and have particular expertise in being able to identify epitopes shared in populations as well as shared across related infections such as Zika, Dengue and West Nile. We have also worked on projects to identify shared epitopes across infections with many serovars, such as Leptospirosis. Our assay and methods can be applied to any exposure, and we have applied this to multiple vaccine epitope characterization projects. In previous projects, we have worked with clinical biomarker scientists and therapeutic area scientists involved in vaccine trial design and execution. Based on the draft proposal, we are looking for partners with extensive domain expertise in virology and vaccine development. We have extensive experience with B cell epitope characterization, and would be excited to collaborate with groups who may work with T cell epitope assays and predictive models.
  • Technical area 1: High-throughput biochemical analysis and protein engineering
Terasaki Institute for Biomedical Innovation Chongming Jiang (chongming.jiang@terasaki.org)
Additional: xiling.shen@terasaki.org
Los Angeles, CA The Terasaki Institute for Biomedical Innovation (TIBI) is a non-profit organization focusing on precision medicine and bioengineering. Our team specializes in carrying out research in immunogenic peptides prediction, patient tumor-derived organoids, organs-on-a-chip, and infectious diseases.
We have built one of the most comprehensive and biggest multilevel peptide databases, which involve the mutated peptides (neoantigens), presented peptides, and clinical peptides. We constructed the unique tumor organoid platform to test and validate the immunogenicity of the peptides in the tumor microenvironment in vitro.
We offer: 1) a novel immunomics prediction framework, which imports the gene expression, HLA loss, antigen processing, peptide loading onto major histocompatibility complex (MHC) molecules, neoantigen quality, T cell recognition, and other features by integrating immunomics, proteomics, genomics, quantitative systems immunology/pharmacology, and multiple machine learning/deep learning algorithms.
2) one of the most comprehensive and biggest multilevel cancer peptides databases, which involve the mutated peptides (neoantigens), presented peptides, T cell epitopes, and clinical peptides.
3) a unique patient tumor-derived organoid platform, MicroOrganoSpheres organoid model (https://www.cell.com/cell-stem-cell/fulltext/S1934-5909(22)00159-X?_returnURL=https%3A%2F%2Flinkinghub.elsevier.com%2Fretrieve%2Fpii%2FS193459092200159X%3Fshowall%3Dtrue). Our unique precision-oncology technology that not only guides personalized treatment in <10 days but also retains key components of the TME, which can test and validate the immunogenicity of the candidate T cell epitopes in the tumor microenvironment in vitro, and are compatible with automated workflows for downstream assays.
Strengths and experience: We have one of the most comprehensive and biggest multilevel peptide databases, which involve mutated peptides (neoantigens), presented peptides, and clinical peptides. We constructed the personalized patient tumor-drived organoid model, which can high-throughput test and validate the anti-tumor immune response of the candidate T cell epitopes quickly! We have enrichment experience in applying our platform to colon cancer, breast cancer, melanoma, and other kinds of human tumors. We are looking to create a collaborative, multi-disciplinary team with cutting-edge expertise in technical areas 1 and/or 3. To do this, we are looking for experimental partners in experimentally detecting peptides presentation and binding affinity with MHC molecules, and designing mRNA vaccines, in vivo experimental measurements of the immune response of these T cell epitopes.
  • Technical area 2: Protein modeling toolkit development for antigen design
WuXi Vaccines Justin Bevere (justin.bevere@wuxibiologics.com)
Additional: jason.he@wuxibiologics.com
NJ, PA, Ireland, Germany, China, Singapore WuXi Vaccines is a leading tier 1 CDMO with expertise in CMC/GMP manufacturer of prophylactic vaccines and antibodies. WuXi Vaccines can provide CMC/GMP expertise for recombinant proteins, virus like particles, viral vector vaccines, pDNA/mRNA, toxoid, conjugate, outer membrane vesicle, as well as other unique modalities. Our organization has capacity to submit 150 INDs and 12 BLAs per year and can support clinical trial material from research grade material, and also phase 1 all the way to commercial manufacturing. We are seeking entities that are looking for GMP manufacturing requirements for their novel vaccine antigens.
  • Technical area 3: Translational candidate development and clinical evaluation
Versatope Therapeutics Incorporated Christopher Locher (christopher.locher@versatope.com)
Additional: erin.neff@versatope.com
Lowell, MA Versatope developed a validated vaccine and drug delivery technology platform using outer membrane vesicles (OMVs). The antigen and antibody display platform produces novel vaccines as well as cell-specific drug delivery to improve pharmacodynamic properties, respectively. The first clinical product, a universal influenza vaccine candidate, VT-105 has completed non-clinical proof of concept studies and is now poised for clinical development in 2024. The clinical development of VT-105 is supported by a broad agency announcement (BAA) contract administered by the National Institute of Allergy and Infectious Diseases (NIAID). Versatope has also established a conjugation technology that we have used successfully to develop vaccine candidates for Plasmodium falciparum malaria and Clostridium difficile infections. Additionally, we have a discovery project in oncology using the technology platform for cell-specific targeting with antibody display that delivers payloads of small molecules and therapeutic nucleic acids. Versatope has gained attention in the field of vaccine design due to our unique technology platform that displays endogenously cloned or heterologous antigens. Versatope uses nano-size vesicles that are naturally produced by many Gram-negative bacteria as part of their normal physiology. Versatope has genetically engineered these nano-vesicles to allow for the presentation of a diverse range of antigens in their native protein conformation as well as display of cell-targeting antibody fragments for cell-specific targeting. The adjuvant properties include various pathogen-associated molecular patterns that increase the immune response to co-presented antigens. The stable structure offers protection from degradation and contributes to the vaccine's shelf life and thermostability, including the ability to be lyophilized for storage and transportation at room temperature. The vaccine vesicles can be taken up by antigen-presenting cells (APCs) and subsequently cross-presented to T cells (for cellular immunity) and B cells (for antibody production). Versatope can add to APECx teaming partners through their proprietary nano-vesicle technology platform that display a diverse array of antigens on a single nanovesicle. The new antigen context of the nano-vesicle may not be affected by prior exposure (original antigenic sin) and afford robust immune responses. Versatope has established both the manufacturing process, developed qualified release assays, and demonstrated safety and efficacy that includes low toxicity. The technology platform includes the know-how to genetic engineer our proprietary cell lines that expresses or conjugates heterologous antigens including virus antigens. This modularity allows for the customization of vaccines to target specific pathogens or diseases. In addition to parenteral immunization/administration, Versatope’s vesicles can be administered through mucosal, oral or transdermal routes, leading to the induction of mucosal immunity and long-term immunity. This is particularly important for pathogens that enter the body through mucosal surfaces, such as respiratory or gastrointestinal pathogens. Versatope has a tool box of proprietary and detoxified cell lines available for antigen and antibody display, fusion cloning and display proteins. The technology platform provides versatility in selecting an appropriate strain based on the desired target product profile of the vaccine or therapeutic candidate. Versatope is seeking partners with expertise in supercomputing capabilities for protein modeling to to design and develop novel antigens. We are also seeking partners with expertise in high throughput screening biochemical analysis and protein engineering to select for novel antigens and antibodies.
  • Technical area 3: Translational candidate development and clinical evaluation
Mayo Clinic Michael Barry (mab@mayo.edu)
Additional: moran.teresa@mayo.edu
Rochester, MN Preclinical and Clinical Research in broad fields spanning virology and medicine. Vaccine engineering, preclinical vaccine testing, clinical trials. Mayo Clinic was rated the number one hospital in the United States by US News and World Reports. Mayo Clinic is a field leader in moving viral therapeutics into clinical testing. Structural analysis and antigen design.
  • Technical area 3: Translational candidate development and clinical evaluation
Michigan Technological University Dukka KC (dbkc@mtu.edu)
Additional: dbkc@mtu.edu
Houghton, MI My lab is primarily focused on developing deep learning/AI tools to decipher relationship between protein sequence, structure, function and evolution. Additionally, my lab is also working on developing computational tools to predict pathogens. We have strong background in developing computational tools including AI/ML/DL tools to predict structure and function of proteins. We have also strong background in developing parallel computing algorithms like GPU-ITASSER.
  • Technical area 2: Protein modeling toolkit development for antigen design
University of Michigan Yongqun "Oliver" He (yongqunh@med.umich.edu)
Additional: moemurra@med.umich.edu
Ann Arbor, MI The He lab in the University of Michigan Medical School has been focusing on the research on vaccine informatics. Specifically, Dr. He has been leading the development of the VIOLIN (https://violinet.org), the most comprehensive vaccine knowledgebase and analysis system. VIOLIN has stored >2,500 vaccines against >100 viral infectious diseases and >400 protective viral antigens associated with these vaccines. Each of these vaccines is well annotated with vaccine components, vaccine formulations, vaccine-induced host responses, and experimental evidence. The protective vaccine antigens have been stored in the VIOLIN Protegen protective antigen database (https://violinet.org/protegen) (PMID: 20959289). Using the Protegen antigens as gold standard data, we have developed several vaccine design tools, including original filtering-based Vaxign (PMID: 20671958), Vaxign-ML (PMID: 32096826) machine learning method, and Vaxign2 vaccine design program that combines Vaxign and Vaxign-ML (PMID: 34009334). We are currently enhancing the VIOLIN/Protegen and tools, and are developing a new deep learning-based vaccine design tool. Meanwhile, we are developing ontologies (e.g., the Vaccine Ontology) for knowledge standardization and AI-ready data generation. The He Lab at University of Michigan can provide strong TA2 support. We can provide state-of-the-art vaccine design tools and vaccine design skills. We have been working in the area since 2005. For example, we used our vaccine design methods (Vaxign-ML and Vaxign) to predict protective vaccine antigens for COVID-19 and found that the SARS-CoV-2 S protein and many non-structural proteins are good COVID-19 vaccine candidates (PMID: 32719684). We also developed a structural vaccinology method that keeps the protein surface with intact sequences and structure so that the antibody response will remain, and meanwhile replaces the internal core of the protein 3D structure with predicted immune epitopes so that T cell responses will be enhanced (PMID: 33398234). The immune epitopes can come from other viruses so that such a structure can potentially be used against multiple viruses. A new deep learning-based vaccine design tool is also under development. Another area of our interest is the development of vaccine-related data repositories and ontologies, as demonstrated by the VIOLIN vaccine knowledgebase development. In addition to providing support for potential team partners for vaccine design in a TA1-3 project application(s), we are also interested in applying for a TA2-only project. The University of Michigan has a world-class medical school, outstanding computing resource, and talented researchers and scientists. Our He Lab has been one of pioneer groups in vaccine informatics research. We have been actively developing popular vaccine design tools and web programs (including Vaxign, Vaxign-ML, and Vaxign2) for over ten years. We have also developed the gold standard protective vaccine antigen database (Protegen), which has been used by many vaccine design programs as training data for vaccine design tool development. Each protective vaccine antigen in the database has been experimentally verified and manually annotated. Such a resource is very worthwhile and can be further developed for more thorough data collection and annotation, which can be a valuable resource for APECx study. We have also developed ontologies including the Vaccine Ontology (VO) and Ontology of Adverse Events (OAE), which have been used as community resources to support vaccine data standardization and analysis. Such ontological representation and standardization would support systematical APECx study and AI research. Furthermore, we have a long history of developing web applications and data repositories. We plan to submit a TA2 only proposal. Meanwhile, we are also interested in providing TA2 vaccine design support for potential teaming partners for TA1-3 application(s). We seek potential teaming partners in either of these two options.
  • Technical area 2: Protein modeling toolkit development for antigen design
Rubix LS Stacy Arrazcaeta (sarrazcaeta@rubixls.com)
Additional: rswift@rubixls.com
Lawrence, MA At Rubix LS, our research is strategically focused on addressing the pressing healthcare challenges faced by underserved and underrepresented communities. With a deep commitment to innovation and inclusivity, our current research encompasses two primary areas: neuroscience and women's health.

In the realm of neuroscience, we are pioneering studies to explore and treat a range of neurological disorders that disproportionately affect diverse populations. Our approach is not just about developing treatments; it's about understanding the nuances of these conditions in different demographic groups. This focus allows us to tailor therapies that are both effective and sensitive to the unique needs of each patient.

Simultaneously, our research in women's health is breaking new ground. We recognize that women, particularly those from minority backgrounds, have historically been underrepresented in medical research. Our goal is to change this narrative by delving into health issues specifically affecting women, from reproductive health to conditions that manifest differently in women. This includes leveraging our extensive patient data to uncover insights and drive the development of personalized treatments.

Overall, Rubix LS is dedicated to transforming healthcare research by ensuring that it is patient-centered, data-driven, and truly inclusive. Our vision is to create a healthcare landscape where every individual, regardless of their background, has access to the best possible care
Rubix LS, with its expertise in leveraging extensive, diverse patient data sets and innovative approach in healthcare research, can significantly contribute to the ARPA-H APECx program. Our organization specializes in developing personalized healthcare solutions, particularly for underserved communities, which aligns with APECx's goal of transforming vaccine discovery. Rubix LS's capability in high-throughput biochemical analysis and protein engineering, combined with our advanced protein modeling toolkit for antigen design, positions us as a valuable partner in this initiative. Our experience in translational candidate development and clinical evaluation can accelerate the development of vaccines targeting entire viral families, fundamentally shifting the vaccine development paradigm towards more efficient and inclusive solutions. Rubix LS specializes in pioneering healthcare research, particularly for underrepresented communities. Our strengths lie in our extensive, diverse patient data sets and our focus on personalized healthcare solutions. We excel in neuroscience and women's health research, leveraging cutting-edge techniques to develop targeted treatments. Our commitment to inclusivity and patient-centered approaches sets us apart in the healthcare sector, enabling us to create innovative, effective healthcare solutions. Rubix LS seeks partners who share our dedication to innovative healthcare solutions, especially for underrepresented communities. Ideal partners would bring expertise in areas like advanced data analytics, clinical research, and patient-centric design. We value collaborators who are committed to breaking new ground in healthcare, through pioneering research methods, and who have a proven track record in translational science, facilitating the journey from research to real-world application. Partners who can enhance our mission with complementary skills and a shared vision for equitable healthcare innovation would be ideal.
  • Technical area 3: Translational candidate development and clinical evaluation