{"id":605209,"date":"2026-08-26T14:00:21","date_gmt":"2026-08-26T14:00:21","guid":{"rendered":"https:\/\/www.newsbeep.com\/ie\/605209\/"},"modified":"2026-08-26T14:00:21","modified_gmt":"2026-08-26T14:00:21","slug":"how-machine-learning-can-uncover-immune-system-mysteries","status":"publish","type":"post","link":"https:\/\/www.newsbeep.com\/ie\/605209\/","title":{"rendered":"How Machine Learning Can Uncover Immune System Mysteries"},"content":{"rendered":"

Original story from Cold Spring Harbor Laboratory (NY, USA).<\/a><\/p>\n

Could AI hold the key to answering questions that have stumped doctors and scientists for decades? A recent study at Cold Spring Harbor Laboratory<\/a> (NY, USA) borrows concepts from machine learning to address an age-old riddle of immunology.<\/p>\n

In the thymus, the immune system\u2019s T cells are trained to avoid attacking healthy tissue via a process called negative selection. There, T cells are tested on whether they bind to fragments of the body\u2019s own proteins, called self-peptides. Those that do are immediately deleted. However, each T cell encounters only a small fraction of the enormous number of self-peptides abounding throughout the body. So, how does the immune system learn to tolerate the rest?<\/p>\n

\u201cThis has long been an open question in immunology,\u201d explained Hannah Meyer. \u201cNegative selection is a crucial process, but if T cells had to test against every single one of the body\u2019s peptides, it would take forever. So, how do they learn to avoid friendly fire? We think it\u2019s through a process called generalization.\u201d<\/p>\n

To anyone who\u2019s worked in tech during the last 10 or so years, that word should ring a bell. \u201cMachine learning deals with the generalization problem all the time,\u201d Saket Navlakha explained. For example, to build a model that can detect dogs, you don\u2019t need to train it on every single dog image on the Web. \u201cMachine learning tells us that generalization is possible, but only under certain conditions. It\u2019s not black magic. So, let\u2019s try to look at this immunology question from that perspective.\u201d<\/p>\n

\"\"<\/a><\/p>\n

World-first AI-designed genomes generate 16 powerful new bacteria-killing viruses<\/a><\/p>\n

In a world-first, scientists have used genome language models to generate new viruses that selectively target bacteria, raising hopes for new medicines.<\/p>\n

Meyer, Navlakha and their team found that the immune system fulfills two key conditions necessary to learn through generalization. First, the abundance of self-peptides in the thymus closely mirrors their abundance in tissues throughout the body. In machine learning terms, the training data resemble the test data. Second, T-cell receptors are cross-reactive, meaning a single receptor can recognize several similar peptides, allowing the T cell to learn about peptides it never directly encounters.<\/p>\n

The study shows that through this process, 90% of self-reactive T cells can be correctly deleted in the thymus despite each one encountering only 10% of the body\u2019s self-peptides. Thus, the team\u2019s research suggests our immune system accomplishes the impressive feat of negative selection through a biological kind of \u2018generalization\u2019.<\/p>\n

To take their findings one important step further, the team asked whether failures of this generalization process could help explain autoimmunity \u2013 when the immune system wrongly attacks healthy tissue. Remarkably, the team\u2019s AI model accurately reproduced features of autoimmune polyendocrine syndrome type 1, a rare autoimmune disease.<\/p>\n

\u201cWe\u2019re calling this direction ImmunoAI,\u201d Navlakha noted. \u201cWe\u2019re not trying to create AI inspired by the immune system, but we\u2019re studying how the immune system solves fundamental machine learning problems. When we start to look at the immune system as if it\u2019s another kind of AI, we may find out some surprising things about human health and disease.\u201d<\/p>\n

This article has been republished from the following\u00a0materials<\/a>.\u00a0Material may have been edited for length and\u00a0house style. For further information, please contact the cited source. Our press release publishing policy can be accessed\u00a0here<\/a>.<\/p>\n","protected":false},"excerpt":{"rendered":"Original story from Cold Spring Harbor Laboratory (NY, USA). Could AI hold the key to answering questions that…\n","protected":false},"author":2,"featured_media":605210,"comment_status":"","ping_status":"","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[10],"tags":[220,79901,103,61,60,3045],"class_list":["post-605209","post","type-post","status-publish","format-standard","has-post-thumbnail","category-health","tag-ai","tag-autoimmunity","tag-health","tag-ie","tag-ireland","tag-machine-learning"],"_links":{"self":[{"href":"https:\/\/www.newsbeep.com\/ie\/wp-json\/wp\/v2\/posts\/605209","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.newsbeep.com\/ie\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.newsbeep.com\/ie\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.newsbeep.com\/ie\/wp-json\/wp\/v2\/users\/2"}],"replies":[{"embeddable":true,"href":"https:\/\/www.newsbeep.com\/ie\/wp-json\/wp\/v2\/comments?post=605209"}],"version-history":[{"count":0,"href":"https:\/\/www.newsbeep.com\/ie\/wp-json\/wp\/v2\/posts\/605209\/revisions"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/www.newsbeep.com\/ie\/wp-json\/wp\/v2\/media\/605210"}],"wp:attachment":[{"href":"https:\/\/www.newsbeep.com\/ie\/wp-json\/wp\/v2\/media?parent=605209"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.newsbeep.com\/ie\/wp-json\/wp\/v2\/categories?post=605209"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.newsbeep.com\/ie\/wp-json\/wp\/v2\/tags?post=605209"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}