A novel artificial intelligence system can access information stored in the immune system about past infections, diseases, and vaccinations and could be harnessed as a clinical tool for medical diagnosis.
The machine learning for immunological diagnosis (Mal-ID) framework, described in the journal Science, could one day lead to a single blood test to diagnose multiple diseases.
The system combines DNA sequencing and machine learning to create a new type of data-driven medical diagnostic that reads the immune system’s record of exposure to disease. Specifically, it captures the vast quantity of information contained in the genetic sequences of B and T immune cell receptors and could act as biomarkers for past immune system activity.
“Our study shows it’s possible to unlock the hidden information in immune receptor sequences in a robust way for many different types of diseases and immune states,” lead researcher Maxim Zaslavsky, PhD, who runs a team at Stanford University, told Inside Precision Medicine.
“We believe this approach may one day be able to help doctors diagnose and treat autoimmune diseases, just like next-generation genomic sequencing transformed cancer care by matching patients to targeted therapies based on their tumor’s genetic profile.”
Clinical diagnosis usually involves physical examinations, a patient’s medical history, laboratory tests, and imaging studies and can result in laborious investigations that may not result in definitive answers, particularly for autoimmune diseases. It also makes little use of the immune system’s own record of exposure to antigens on pathogens and even the body’s own tissues that are reflected in the B and T cell receptors of the immune system.
B and T cell receptor populations change after exposure to pathogens, after vaccination, and in response to autoantigens in autoimmune conditions, which reflects clonal expansion and selection of B cells and T cells during immune responses.
Genes encoding these cells are generated by a random recombination of segments in the genome of individual cells during their development and could potentially represent a diverse set of sequence biomarkers associated with immune system activity.
Zaslavsky and co-workers created the Mal-ID method for identifying B cell receptor heavy chain and T cell receptor beta chain features characteristic of infectious and immunological disorders or generated by therapeutic or prophylactic interventions such as vaccination.
Mal-ID combines traditional immunological analyses such as the detection of shared sequences between people who have the same condition, with more complex features derived from AI models of protein sequences, called protein language models.
While AI systems can be difficult to interpret, the team developed ways to understand how the model made its diagnostic predictions.
Mal-ID accurately identified the immune status of blood samples from 542 people with COVID-19, HIV, lupus, Type 1 diabetes, recent flu vaccination, and healthy individuals, with a multiclass area under the receiver operating characteristic curve (AUROC) of 0.986 on data not used for training.
Combining features from both B cell and T cell receptor data performed best but even with only B cell receptor sequences it achieved an AUROC of 0.959 using an expanded cohort that added 51 individuals for whom only these data were available.
Zaslavsky said: “What we have today is a proof of concept and requires further validation, but this approach of directly measuring the immune cells that may be driving the diseases could give us deeper insight into the underlying mechanisms of disease, especially for diseases that might take years to diagnose and where patients have to endure an agonizing and expensive trial-and-error process to find the right treatment.”
Website: International Conference on Infectious Diseases
#InfectiousDiseases, #ID2024, or relevant year, #GlobalHealth, #DiseasePrevention, #InfectiousDiseaseConference, #PublicHealth, #Epidemiology, #DiseaseControl, #HealthInnovation, #VaccinationMatters, #PandemicPreparedness, #PathogenResearch, #OneHealth, #InfectionPrevention, #GlobalDiseaseOutbreak, #ViralResearch, #EmergingInfectiousDiseases, #HealthSecurity, #MedicalConference, #ICIDSummit
Visit Our Website : infectious-diseases-conferences.pencis.com
Nomination Link : infectious-diseases-conferences.pencis.com/award-nomination
Registration Link : infectious-diseases-conferences.pencis.com/award-registration
Member Link : infectious-diseases-conferences.pencis.com/conference-membership
Awards-Winners : infectious-diseases-conferences.pencis.com/awards-winners/
Contact us : infectious@pencis.com
Get Connected Here:
==================
Social Media Link
Twitter : twitter.com/skyla00827177
Blog : infectious2021.blogspot.com
Instagram : www.instagram.com/infectious_diseases2021
Facebook : www.facebook.com/pencis.queen
YouTube : www.youtube.com/channel
LinkedIn : www.linkedin.com/in/infectious-diseases-conferences
Wikipedia : infectiousdiseases625904580.wordpress.com
No comments:
Post a Comment