{"id":590287,"date":"2026-08-11T20:46:08","date_gmt":"2026-08-11T20:46:08","guid":{"rendered":"https:\/\/www.newsbeep.com\/ie\/590287\/"},"modified":"2026-08-11T20:46:08","modified_gmt":"2026-08-11T20:46:08","slug":"new-machine-learning-model-improves-accuracy-of-prenatal-genetic-testing","status":"publish","type":"post","link":"https:\/\/www.newsbeep.com\/ie\/590287\/","title":{"rendered":"New machine learning model improves accuracy of prenatal genetic testing"},"content":{"rendered":"<p>Advances in genome sequencing\u00a0are giving\u00a0more\u00a0families access to\u00a0prenatal\u00a0genetic\u00a0testing and\u00a0new information\u00a0about an\u00a0unborn baby&#8217;s\u00a0health,\u00a0including\u00a0whether genetic changes may\u00a0link to\u00a0a\u00a0neurodevelopmental\u00a0condition.\u00a0<\/p>\n<p>But testing can also uncover DNA changes that clinicians do not yet understand. Known as\u00a0variants of uncertain significance\u00a0or\u00a0VUS,\u00a0these changes\u00a0can&#8217;t\u00a0be easily classified\u00a0as\u00a0harmful or harmless\u00a0based on available evidence.\u00a0The result:\u00a0uncertainty\u00a0for a\u00a0significant number\u00a0of\u00a0expecting parents.\u00a0<\/p>\n<p>&#13;<\/p>\n<p>Doctors\u00a0may\u00a0detect a\u00a0new\u00a0genetic change, but in about one-third of\u00a0cases they won&#8217;t know if it&#8217;s truly\u00a0abnormal or\u00a0simply\u00a0within the normal sphere\u00a0of development. With\u00a0too many variants for geneticists to keep up with,\u00a0it&#8217;s impossible for them to\u00a0always know\u00a0if\u00a0a variant of\u00a0interest may\u00a0result in\u00a0neurodevelopmental or other genetic disorders.&#8221;\u00a0<\/p>\n<p>&#13;<br \/>\n&#13;<\/p>\n<p style=\"text-align: right;\">Dr. Rosanna Weksberg, Clinical Geneticist and Senior Associate Scientist, Genetics &amp; Genome Biology at The Hospital for Sick Children (SickKids)<\/p>\n<p>&#13;<\/p>\n<p>Members of the\u00a0Weksberg\u00a0lab\u00a0are\u00a0experts\u00a0in <a href=\"https:\/\/www.news-medical.net\/health\/Understanding-The-Role-of-Epigenetics-in-Human-Disease.aspx\" class=\"linked-term\" rel=\"nofollow noopener\" target=\"_blank\">epigenetics<\/a>,\u00a0the study of\u00a0factors that influence how genes\u00a0are turned on and off.\u00a0Their newest innovation is a machine learning model that can help clinicians assess uncertain variants no matter what samples they are able to use and, as a result, help bring long-needed clarity to families.\u00a0<\/p>\n<p>Transforming the use of &#8216;episignatures&#8217;\u00a0<\/p>\n<p>Episignatures\u00a0are\u00a0distinct\u00a0chemical tags in\u00a0DNA\u00a0that\u00a0reveal the presence of specific genetic\u00a0conditions.\u00a0In\u00a02020, the team\u00a0unveiled a platform,\u00a0EpigenCentral,\u00a0that\u00a0uses blood-derived\u00a0episignature\u00a0data\u00a0to help\u00a0doctors\u00a0determine\u00a0if a variant is\u00a0disease-causing\u00a0or benign.\u00a0<\/p>\n<p>To date, the Weksberg lab has helped\u00a0establish\u00a0more than\u00a060\u00a0episignatures,\u00a0most\u00a0clinically proven for making diagnoses.\u00a0Until now, these patterns\u00a0have been\u00a0tissue-specific,\u00a0meaning that\u00a0one found in blood could\u00a0not\u00a0be applied to\u00a0other\u00a0tissues, leaving this\u00a0valuable diagnostic approach out of reach when\u00a0testing\u00a0amniotic fluid or placental tissue\u00a0in prenatal settings.\u00a0<\/p>\n<p>&#8220;If we could take these blood-derived signatures and make them tissue-agnostic, we could overcome one of the biggest limitations in prenatal diagnostics,&#8221;\u00a0says Dr. Sanaa Choufani,\u00a0Senior\u00a0Research Associate.\u00a0<\/p>\n<p>This inspired their\u00a0new\u00a0machine\u00a0learning\u00a0method, which\u00a0could\u00a0transform blood-derived\u00a0episignatures\u00a0into &#8220;tissue-agnostic&#8221;\u00a0ones\u00a0(those that stay the same no matter where they originated in the body).\u00a0That\u00a0is, they are informative in multiple tissue types including\u00a0prenatal\u00a0samples.\u00a0Weksberg and Choufani co-led\u00a0a\u00a0new study\u00a0in\u00a0The American Journal of Human Genetics\u00a0demonstrating that this approach indeed works,\u00a0opening the door to\u00a0its\u00a0future diagnostic use in prenatal medicine.\u00a0<\/p>\n<p>New model\u00a0proves effective in\u00a0identifying\u00a0variants\u00a0<\/p>\n<p>In proving the\u00a0concept,\u00a0the team generated a blood-derived\u00a0episignature\u00a0using samples from\u00a0266\u00a0people with\u00a0Down\u00a0syndrome, selected because it is one of the more common rare genetic conditions.\u00a0They then trained\u00a0their\u00a0model using publicly available DNA\u00a0methylation\u00a0data\u00a0(chemical tags)\u00a0from 850 individuals with and without Down\u00a0syndrome, covering\u00a0six different prenatal and postnatal tissue types.\u00a0<\/p>\n<p>Their\u00a0model accurately\u00a0recognized the\u00a0Down\u00a0syndrome\u00a0pattern\u00a0in\u00a0every\u00a0tissue\u00a0type tested,\u00a0showing that\u00a0a\u00a0blood-derived\u00a0episignature\u00a0can be changed into\u00a0one\u00a0capable of\u00a0identifying\u00a0a\u00a0disease-specific\u00a0episignature\u00a0across\u00a0different\u00a0tissues.\u00a0<\/p>\n<p>&#8220;We&#8217;re thrilled that our model can bring a new level of precision to prenatal testing, where\u00a0so many\u00a0questions\u00a0remain to be answered,&#8221; Weksberg says. &#8220;This machine\u00a0learning\u00a0approach is also a\u00a0building block\u00a0to study many disorders where tissues are inaccessible,\u00a0which would\u00a0support rapid translation\u00a0into the clinic.&#8221;\u00a0<\/p>\n<p>In fact, the researchers\u00a0say\u00a0this\u00a0could eventually support diagnostic testing using\u00a0a broader range of\u00a0samples, such as\u00a0saliva and oral swabs\u00a0&#8211;\u00a0thereby reducing the need for blood or\u00a0other\u00a0tissue.\u00a0<\/p>\n<p>In\u00a0overcoming\u00a0a\u00a0key limitation of epigenetic testing, the\u00a0team\u00a0hopes to\u00a0one day\u00a0enable earlier, more\u00a0accurate\u00a0diagnoses\u00a0and\u00a0help families make decisions\u00a0with\u00a0clearer information. The work supports\u00a0Precision Child Health,\u00a0a movement at SickKids to\u00a0support\u00a0individualized\u00a0care\u00a0for each child,\u00a0including\u00a0better diagnosis and prediction.\u00a0<\/p>\n<p>&#8220;When families are told of a potential neurodevelopmental disorder, our goal with this approach is to provide more information, as detailed as possible, to support them. We want to reduce the diagnostic odyssey that many of these patients go through for rare disorders,&#8221; Choufani says.\u00a0<\/p>\n<p>Source:<\/p>\n<p><a href=\"https:\/\/www.sickkids.ca\/en\/news\/archive\/2026\/machine-learning-approach-could-bring-greater-certainty-to-prenatal-genetic-testing\/\" rel=\"noopener nofollow\" target=\"_blank\">The Hospital for Sick Children<\/a><\/p>\n","protected":false},"excerpt":{"rendered":"Advances in genome sequencing\u00a0are giving\u00a0more\u00a0families access to\u00a0prenatal\u00a0genetic\u00a0testing and\u00a0new information\u00a0about an\u00a0unborn baby&#8217;s\u00a0health,\u00a0including\u00a0whether genetic changes may\u00a0link to\u00a0a\u00a0neurodevelopmental\u00a0condition.\u00a0 But testing can&hellip;\n","protected":false},"author":2,"featured_media":2704,"comment_status":"","ping_status":"","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[10],"tags":[83,2991,4103,8016,619,3302,254,3303,103,3175,61,60,3045,13252,89,1246],"class_list":["post-590287","post","type-post","status-publish","format-standard","has-post-thumbnail","category-health","tag-baby","tag-blood","tag-children","tag-diagnostic","tag-dna","tag-genetic","tag-genetics","tag-genome","tag-health","tag-hospital","tag-ie","tag-ireland","tag-machine-learning","tag-prenatal","tag-research","tag-syndrome"],"_links":{"self":[{"href":"https:\/\/www.newsbeep.com\/ie\/wp-json\/wp\/v2\/posts\/590287","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=590287"}],"version-history":[{"count":0,"href":"https:\/\/www.newsbeep.com\/ie\/wp-json\/wp\/v2\/posts\/590287\/revisions"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/www.newsbeep.com\/ie\/wp-json\/wp\/v2\/media\/2704"}],"wp:attachment":[{"href":"https:\/\/www.newsbeep.com\/ie\/wp-json\/wp\/v2\/media?parent=590287"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.newsbeep.com\/ie\/wp-json\/wp\/v2\/categories?post=590287"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.newsbeep.com\/ie\/wp-json\/wp\/v2\/tags?post=590287"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}