Researchers from the Pacific Northwest National Laboratory (PNNL) and Harvard Medical School (HMS) are utilizing artificial intelligence (AI) to expedite vaccine development. Through their Rapid Assessment of Platform Technologies to Expedite Response (RAPTER) project, the scientists are leveraging machine learning and AI to search scientific literature for knowledge on constructing effective vaccines against new infectious viruses and bacteria.
Traditionally, vaccine development is a time-consuming and costly process, often taking years and millions of dollars to complete. Vaccines are typically created using different strategies known as platforms, which can generate distinct immune responses. The RAPTER tool aims to determine the most effective strategy for a specific virus or bacteria, streamlining vaccine production and reducing the timeline and cost.
To deal with the overwhelming amount of information in scientific literature, the researchers are developing RAPTER to automatically sift through publications and catalog results from various vaccine design experiments. This will provide decision-makers with valuable insights for selecting the best strategy during future pandemics.
The collaboration between PNNL and HMS involves extracting meaningful information from scientific publications. While HMS has already developed similar tools for small molecule design, vaccines and immunology present greater complexities. However, by building upon existing tools, the researchers aim to extract key information from publications and gain a deeper understanding of immune responses using different vaccine strategies.
Key terms that connect mechanisms of immunity to experimental measurements are defined to enable the RAPTER tool to identify relationships between terms across various scientific publications. This information is then used to construct an extensive graph of relationships within the immune response, aiding in predicting the effectiveness of different vaccine strategies.
The Defense Threat Reduction Agency (DTRA), responsible for mitigating emerging threats, supports the RAPTER project to defend against future pandemics. The project comprises a consortium of research institutes, led by Los Alamos National Laboratory (LANL), which collects and curates raw experimental data on viruses and vaccines. Artificial intelligence is employed to identify patterns within the data for each vaccine candidate. Collaborating institutes, including Lawrence Livermore National Laboratory, Sandia National Laboratories, and various universities, contribute to the project.
Once the computational tools are developed, the research institutes will unite their efforts to experimentally validate the results. Confirming the computational findings for the mRNA platform, the same platform used in the vaccine against SARS-CoV-2, will be carried out by researchers such as Zachary Stromberg from PNNL.
By automating the vaccine design process, it is hoped that scientists will be able to expedite their efforts and develop vaccines more efficiently. This collaboration between researchers, aided by AI, offers promising advancements in the fight against future pandemics.
