{"paper_id":"07347f7b-a55d-403d-bafd-8d716ebd3ca4","body_text":"1 \n \n\u0018ĞĮŶŝŶŐ\u0003ƚŚĞ\u0003\u0018E\u0004\u0003\u0011ŝŶĚŝŶŐ\u0003^ƉĞĐŝĮĐŝƚLJ\u0003ŽĨ\u0003'Z,>Ϯ\u0003\u0003\nPaige E. Messa1†, Christopher L. Warren1†, Noah R. Nicol1, Keenan S. Pearson1͕\u0003:ƵƐƟŶ\u0003W ͘\u0003WĞƚĞƌƐ2, \nAmy M. Fowler3,4,5, Elaine T. Alarid6, Mary Szatkowski Ozers1,3,* \n1Proteovista LLC, Madison, WI 53719, USA \n2Department of Chemistry and Biochemistry, University of Northern Iowa, Cedar Falls, IA 50614, \nUSA \n3Department of Radiology, University of Wisconsin-Madison, Madison, WI 53792, USA \n4Department of Medical Physics, University of Wisconsin-Madison, Madison, WI 53705, USA \n5Carbone Comprehensive Cancer Center, University of Wisconsin-Madison, Madison, WI 53705, \nUSA \n6McArdle Laboratory for Cancer Research, Department of Oncology, Carbone Comprehensive \nCancer Center, University of Wisconsin-Madison, Madison, WI 53705, USA \n†Paige E. Messa1 and Christopher L. Warren1 contributed equally to this work.  \n*To whom correspondence should be addressed. Email: mozers@proteovista.com \n  \n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted April 18, 2026. ; https://doi.org/10.64898/2026.04.16.719077doi: bioRxiv preprint \n\n2 \n \n\u0004\u0011^dZ\u0004\u0012d\u0003\nGrainyhead-ůŝŬĞ\u0003Ϯ\u0003;'Z,>ϮͿ\u0003ŝƐ\u0003ĂŶ\u0003ĞƉŝƚŚĞůŝĂů\u0003ƚƌĂŶƐĐƌŝƉƟŽŶ\u0003ĨĂĐƚŽƌ\u0003ǁŝƚŚ\u0003ĐŽŶƚĞdžƚ-dependent \nƌĞŐƵůĂƚŽƌLJ\u0003ƌŽůĞƐ͕\u0003LJĞƚ\u0003ƚŚĞ\u0003ƐĞƋƵĞŶĐĞ\u0003ƌƵůĞƐ\u0003ŐŽǀĞƌŶŝŶŐ\u0003ŝƚƐ\u0003\u0018E\u0004\u0003ƌĞĐŽŐŶŝƟŽŶ\u0003ƌĞŵĂŝŶ\u0003ŝŶĐŽŵƉůĞƚĞůLJ\u0003\nĚĞĮŶĞĚ͘\u0003/Ŷ\u0003ƚŚŝƐ\u0003ƐƚƵĚLJ͕\u0003Ă\u0003ŚŝŐŚ-density genomic ^ƉĞĐŝĮĐŝƚLJ\u0003ĂŶĚ\u0003\u0004ĸŶŝƚLJ\u0003ĨŽƌ\u0003WƌŽƚĞŝŶ\u0003;SNAP) DNA-\nďŝŶĚŝŶŐ\u0003ĂƌƌĂLJ\u0003ĐŽŶƚĂŝŶŝŶŐ\u0003ϳϳϮ͕ϳϯϮ\u0003ƟůĞĚ\u0003ƉƌŽďĞƐ\u0003ĚĞƌŝǀĞĚ\u0003ĨƌŽŵ\u0003'Z,>Ϯ\u0003\u0012Ś/W-seq regions was used to \nƌĞƐŽůǀĞ\u0003'Z,>Ϯ\u0003ďŝŶĚŝŶŐ\u0003ƐƉĞĐŝĮĐŝƚLJ\u0003Ăƚ\u00036 base pair ƌĞƐŽůƵƟŽŶ\u0003ĂĐƌŽƐƐ\u0003ŐĞŶŽŵŝĐ\u0003ƐĞƋƵĞŶĐĞƐ͘\u0003From \nhigh-ĂĸŶŝƚLJ\u0003ƉƌŽďĞƐ͕\u0003ĚĞ\u0003ŶŽǀŽ\u0003ŵŽƟĨ\u0003ĂŶĂůLJƐŝƐ\u0003ƌĞĐŽǀĞƌĞĚ\u0003ƚŚĞ\u0003ĐĂŶŽŶical 5’-AACCGGTT-ϯ͛\u0003ŵŽƟĨ͘\u0003\n^ĞƋƵĞŶĐĞ\u0003ƐƉĞĐŝĮĐŝƚLJ\u0003ůĂŶĚƐĐĂƉĞƐ\u0003ƌĞǀĞĂůĞĚ\u0003Ă\u0003ƐƚĞƉǁŝƐĞ\u0003ƌĞĚƵĐƟŽŶ\u0003ŝŶ\u0003ďŝŶĚŝŶŐ\u0003ĂƐ\u0003ŵŝƐŵĂƚĐŚĞƐ\u0003ǁĞƌĞ\u0003\nŝŶƚƌŽĚƵĐĞĚ͕\u0003ǁŝƚŚ\u0003ƚŚĞ\u0003ƐƚƌŽŶŐĞƐƚ\u0003ĞīĞĐƚƐ\u0003Ăƚ\u0003ƚŚĞ\u0003\u0012\u0003;ƉŽƐŝƟŽŶ\u0003ϯͿ\u0003ĂŶĚ\u0003'\u0003;ƉŽƐŝƟŽŶ\u0003ϲͿ\u0003ǁŝƚŚŝŶ\u0003ƚŚĞ\u0003ŵŽƟĨ͕\u0003\ngreater tolerance at the central CG dinƵĐůĞŽƟĚĞ͕\u0003ĂŶĚ\u0003ŝŶƚĞƌŵĞĚŝĂƚĞ\u0003ƚŽůĞƌĂŶĐĞ\u0003Ăƚ\u0003ƚŚĞ\u0003\u0004ͬd\u0003ďĂƐĞƐ\u0003Ăƚ\u0003\nƚŚĞ\u0003ŵŽƟĨ\u0003ĞĚŐĞƐ͘\u0003dŚŝƐ\u0003ĂŶĂůLJƐŝƐ\u0003ĂůƐŽ\u0003ĚĞŵŽŶƐƚƌĂƚĞĚ\u0003ƚŚĞ\u0003ŝŶŇƵĞŶĐĞ\u0003ŽĨ\u0003ŶĞĂƌďLJ\u0003ŇĂŶŬŝŶŐ\u0003ƐĞƋƵĞŶĐĞƐ͘\u0003\n\u001cdžƚĞŶĚĞĚ\u0003ŵŽƟĨ\u0003ĂŶĚ\u0003ƐƉĂĐŝŶŐ\u0003ĂŶĂůLJƐĞƐ\u0003ŝŶĚŝĐĂƚĞĚ\u0003ĚŝŵĞƌŝĐ\u0003ďŝŶĚŝŶŐ\u0003Ăƚ\u0003ƉĂŝƌĞĚ\u0003ŵŽƟĨƐ͕\u0003ǁŝƚŚ\u0003ƉĞƌŝŽĚŝĐ\u0003\nhelical spaĐŝŶŐ\u0003ĐŽŶƐŝƐƚĞŶƚ\u0003ǁŝƚŚ\u0003ŝŶƚĞƌĂĐƟŽŶƐ\u0003ŽŶ\u0003ƚŚĞ\u0003ƐĂŵĞ\u0003ĨĂĐĞ\u0003ŽĨ\u0003ƚŚĞ\u0003\u0018E\u0004\u0003ŚĞůŝdž͘\u0003/ŶƚĞŐƌĂƟŽŶ\u0003ŽĨ\u0003\nSNAP array binding with ChIP-ƐĞƋ\u0003ĚĂƚĂ\u0003ĚŝƐƟŶŐƵŝƐŚĞĚ\u0003ĚŝƌĞĐƚ͕\u0003ŵŽƟĨ-encoded GRHL2 occupancy \nfrom indirect, cofactor-mediated recruitment at genomic sites. These ƌĞƐƵůƚƐ\u0003ĚĞĮŶĞ\u0003ƚŚĞ\u0003\nƐĞƋƵĞŶĐĞ\u0003ƐƉĞĐŝĮĐŝƚLJ\u0003ŽĨ\u0003'Z,>Ϯ\u0003ŝŶƚĞƌĂĐƟŽŶƐ\u0003ǁŝƚŚ\u0003ǀĂƌŝĂƟŽŶƐ\u0003ŝŶ\u0003ƚŚĞ\u0003\u0018E\u0004\u0003ĐŽŶƐĞŶƐƵƐ\u0003ŵŽƟĨ\u0003ĂŶĚ\u0003\nŇĂŶŬŝŶŐ\u0003ƐĞƋƵĞŶĐĞƐ\u0003ǁŝƚŚŝŶ\u0003ĂŶ\u0003ĞŶĚŽŐĞŶŽƵƐ\u0003ŐĞŶŽŵŝĐ\u0003ĐŽŶƚĞdžƚ͘  \n  \n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted April 18, 2026. ; https://doi.org/10.64898/2026.04.16.719077doi: bioRxiv preprint \n\n3 \n \n/EdZK\u0018h\u0012d/KE\u0003\n The Grainyhead-like (GRHL) protein family is a highly conserved family of nuclear \nƚƌĂŶƐĐƌŝƉƟŽŶ\u0003ĨĂĐƚŽƌƐ\u0003ĐŽŵƉƌŝƐĞĚ\u0003ŽĨ\u0003'Z,>ϭ͕\u0003'Z,>Ϯ͕\u0003ĂŶĚ\u0003'Z,>ϯ͕\u0003ĞĂĐŚ\u0003ŽĨ\u0003ǁŚŝĐŚ\u0003ƉůĂLJ\u0003ǀŝƚĂů\u0003ƌŽůĞƐ\u0003ŝŶ\u0003\nepithelial development and maintenance (1-4). Although GRHL family members share structural \nŚŽŵŽůŽŐLJ͕\u0003 ŝŶĐůƵĚŝŶŐ\u0003 ƌĞĐŽŐŶŝƟŽŶ\u0003 ŽĨ\u0003 ĂŶ\u0003 ŝĚĞŶƟĐĂů\u0003 \u0018E\u0004\u0003 ĐŽŶƐĞŶƐƵƐ\u0003 ƐĞƋƵĞŶĐĞ\u0003 ;ϱ͛-AACCGGTT-3’), \nŝŶĚŝǀŝĚƵĂů\u0003ĨĂŵŝůLJ\u0003ŵĞŵďĞƌƐ\u0003ƉƌŽǀŝĚĞ\u0003ĚŝƐƟŶĐƚ\u0003ďŝŽůŽŐŝĐĂů\u0003ĨƵŶĐƟŽŶƐ\u0003ĂŶĚ\u0003ƉŽƐƐĞƐƐ\u0003ƐƉĂƟĂů\u0003ĂŶĚ\u0003ƚĞŵƉŽƌĂů\u0003\nĚŝīĞƌĞŶĐĞƐ\u0003ŝŶ\u0003ĞdžƉƌĞƐƐion during developmental stages (1). Loss of GRHL2 in mice causes severe \nĚĞĨĞĐƚƐ\u0003ŝŶ\u0003ŶĞƵƌĂů\u0003ƚƵďĞ\u0003ĐůŽƐƵƌĞ\u0003Ăƚ\u0003ĂŶƚĞƌŝŽƌ\u0003ĂŶĚ\u0003ƉŽƐƚĞƌŝŽƌ\u0003ƌĞŐŝŽŶƐ\u0003ƌĞƐƵůƟŶŐ\u0003ŝŶ\u0003ĞŵďƌLJŽŶŝĐ\u0003ůĞƚŚĂůŝƚLJ\u0003\n(5, 6)͘\u0003\u0011ĞLJŽŶĚ\u0003ŝƚƐ\u0003ĚĞǀĞůŽƉŵĞŶƚĂů\u0003ƌŽůĞƐ͕\u0003'Z,>Ϯ\u0003ŝƐ\u0003Ă\u0003ŬĞLJ\u0003ƌĞŐƵůĂƚŽƌ\u0003ŽĨ\u0003ĞƉŝƚŚĞůŝĂů\u0003ĐĞůů\u0003ŝĚĞŶƟƚLJ that has \nbeen shown to ĞdžŚŝďŝƚ\u0003Ă\u0003ŚLJďƌŝĚ\u0003ƌŽůĞ in epithelial-to-ŵĞƐĞŶĐŚLJŵĂů\u0003ƚƌĂŶƐŝƟŽŶ\u0003;\u001cDdͿ͕\u0003ďLJ\u0003ŝŶŚŝďŝƟŶŐ\u0003\nEMT ďƵƚ\u0003ĂůƐŽ\u0003ďLJ\u0003ƉƌŽŵŽƟŶŐ\u0003cellular ŵŝŐƌĂƟŽŶ ĂŶĚ\u0003ƉůĂƐƟĐŝƚLJ\u0003in Ă\u0003ƉƌŽĐĞƐƐ\u0003ĐƌŝƟĐĂů\u0003ŝŶ\u0003ĐĂŶĐĞƌ\u0003ĞƟŽůŽŐLJ \n(3, 4, 7-9).  \n GRHL2 can ĨƵŶĐƟŽŶ\u0003ĂƐ\u0003ďŽƚŚ\u0003Ă\u0003ƚƌĂŶƐĐƌŝƉƟŽŶĂů\u0003ĂĐƟǀĂƚŽƌ\u0003ĂŶĚ\u0003ƌĞƉƌĞƐƐŽr depending on the \nĐĞůůƵůĂƌ\u0003ĐŽŶƚĞdžƚ   (9-11). Genome-wide ChIP-ƐĞƋ\u0003ƐƚƵĚŝĞƐ\u0003ĂĐƌŽƐƐ\u0003ĞƉŝƚŚĞůŝĂů\u0003ĐĞůů\u0003ƚLJƉĞƐ\u0003ŚĂǀĞ\u0003ŝĚĞŶƟĮĞĚ\u0003\nthousands of GRHL2 binding sites, and ƚƌĂŶƐĐƌŝƉƚŽŵŝĐ\u0003 ĂŶĂůLJƐĞƐ\u0003 ĨŽůůŽǁŝŶŐ\u0003 'Z,>Ϯ\u0003 ƉĞƌƚƵƌďĂƟŽŶ\u0003\nĚĞŵŽŶƐƚƌĂƚĞ\u0003 ƌĞŐƵůĂƟŽŶ\u0003 ŽĨ\u0003 ŚƵŶĚƌĞĚƐ\u0003 ŽĨ\u0003 ŐĞŶĞƐ\u0003 ŝŶ\u0003 Ă\u0003 ĐŽŶƚĞdžƚ-dependent manner (12-16). Many \nƚƌĂŶƐĐƌŝƉƟŽŶĂů\u0003ĞīĞĐƚƐ\u0003ĂƉƉĞĂƌ to be indirect, meaning that they are mediated through a protein \nĐŽŵƉůĞdž without direct contacts between GRHL2 and genomic DNA (17, 18)͘\u0003KŶĞ\u0003ƐƵĐŚ\u0003ŝŶƚĞƌĂĐƟŶŐ\u0003\npartner ŚĂƐ\u0003ďĞĞŶ\u0003ŝĚĞŶƟĮĞĚ\u0003ĂƐ estrogen receptor ɲ\u0003;ER) (19) ǁŚŝĐŚ\u0003ŝƐ\u0003ĞdžƉƌĞƐƐĞĚ\u0003ŝŶ\u0003ŽǀĞƌ\u0003ϳϬй\u0003ŽĨ\u0003Ăůů\u0003\nbreast cancers and is Ă\u0003ŵĂũŽƌ\u0003ƚŚĞƌĂƉĞƵƟĐ\u0003ƚĂƌŐĞƚ\u0003(20-23). \n As the most abundant ERɲ\u0003 post-ƚƌĂŶƐůĂƟŽŶĂů\u0003 ŵŽĚŝĮĐĂƟŽŶ͕\u0003 ƉŚŽƐƉŚŽƌLJůĂƟŽŶ\u0003 Ăƚ\u0003 ^ϭϭϴ\u0003 ŝƐ\u0003\nĂƐƐŽĐŝĂƚĞĚ\u0003ǁŝƚŚ\u0003ĞŶŚĂŶĐĞĚ\u0003ĐŚƌŽŵĂƟŶ\u0003ďŝŶĚŝŶŐ\u0003ĂŶĚ\u0003ĞƐƚƌŽŐĞŶ-ƌĞƐƉŽŶƐŝǀĞ\u0003ŐĞŶĞ\u0003ĞdžƉƌĞƐƐŝŽŶ͕\u0003ĂŶĚ\u0003ŝƚƐ\u0003\n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted April 18, 2026. ; https://doi.org/10.64898/2026.04.16.719077doi: bioRxiv preprint \n\n4 \n \npresence has been linked to endocrine therapy outcomes in ERɲ-ƉŽƐŝƟǀĞ\u0003ƚƵŵŽƌƐ (24-28). GRHL2 \nŵŽƟĨƐ\u0003 ĞŶƌŝĐŚĞĚ\u0003 Ăƚ\u0003 ƉŚŽƐƉŚŽƌLJůĂƚĞĚ\u0003ERɲ (pS118-ER ɲ) ďŝŶĚŝŶŐ\u0003 ƐŝƚĞƐ\u0003 ŚĂǀĞ\u0003 ďĞĞŶ\u0003 ŝĚĞŶƟĮĞĚ͕\u0003\nƐƵŐŐĞƐƟŶŐ\u0003Ă\u0003ĐŽŽƉĞƌĂƟŶŐ\u0003ƌŽůĞ\u0003ďĞƚǁĞĞŶ\u0003'Z,>Ϯ\u0003ĂŶĚ\u0003Ɖ^ϭϭϴ-ERɲ ŝŶ\u0003ƌĞŐƵůĂƟŶŐ\u0003ŐĞŶĞ\u0003ĞdžƉƌĞƐƐŝŽŶ\u0003(29). \nSubsequent transcriptomic analysis revealed several genes co-regulated by ER ɲ and GRHL2, as \nevidence of the ability of GRHL2 to both enhance and suppress ER ɲ ƚƌĂŶƐĐƌŝƉƟŽŶĂů\u0003ĂĐƟǀŝƚLJ\u0003(30). \n\u0018E\u0004\u0003 ďŝŶĚŝŶŐ\u0003 ŵŽƟĨ\u0003 ĂŶĂůLJƐŝƐ\u0003 ĚĞŵŽŶƐƚƌĂƚĞĚ\u0003 ƐŝŐŶŝĮĐĂŶƚ\u0003 ĞŶƌŝĐŚŵĞŶƚ\u0003 ŽĨ\u0003the 'Z,>Ϯ\u0003 ŵŽƟĨ\u0003 ŝŶ\u0003 ƚŚĞ\u0003\ntumor-ƐƉĞĐŝĮĐ\u0003ERɲ ĐŝƐƚƌŽŵĞ͕\u0003 ĂŶĚ\u0003 'Z,>Ϯ\u0003 ŝƐ\u0003 ĞƐƐĞŶƟĂů\u0003 ĨŽƌ\u0003 ĞƐƚƌŽŐĞŶ-ƐƟŵƵůĂƚĞĚ\u0003 ƉƌŽůŝĨĞƌĂƟŽŶ\u0003 ŝŶ\u0003\nbreast cancer cells (15)͘\u0003dŚĞƌĞ\u0003ŝƐ\u0003ƐƵďƐƚĂŶƟĂů\u0003ĞǀŝĚĞŶĐĞ\u0003ŽĨ\u0003ƚƵŵŽƌ\u0003ƐƵƉƉƌĞƐƐŝǀĞ\u0003ĂĐƟŽŶ\u0003ďLJ\u0003'Z,>Ϯ\u0003ŝŶ\u0003Ă\u0003\nƟƐƐƵĞ-ƐƉĞĐŝĮĐ\u0003ŵĂŶŶĞƌ\u0003ďLJ\u0003ƉƌĞǀĞŶƟŽŶ\u0003ŽĨ\u0003\u001cDd as well, but GRHL2 ŽǀĞƌĞdžƉƌĞƐƐŝŽŶ is ƉĂƌĂĚŽdžŝĐĂůůLJ\u0003\ncorrelated with poor prognosis clinically (10). ER ɲ\u0003and GRHL2 have also been shown to interact \nǁŝƚŚ\u0003&Ždž\u0004ϭ͕\u0003ǁŚŝĐŚ\u0003ŝƐ\u0003ŽĨ\u0003ŝŶƚĞƌĞƐƚ\u0003ƐŝŶĐĞ\u0003ďŽƚŚ\u0003&Ždž\u0004ϭ\u0003ĂŶĚ\u0003'Z,>Ϯ\u0003ŚĂǀĞ\u0003been shown ƚŽ\u0003ĞdžŚŝďŝƚ\u0003pioneer \nĨĂĐƚŽƌ\u0003ĂĐƟǀŝƚLJ\u0003(19, 31, 32). \n Before GRHL2 ŝŶƚĞƌĂĐƟŽŶƐ\u0003ǁŝƚŚ\u0003these ŽƚŚĞƌ\u0003ƚƌĂŶƐĐƌŝƉƟŽŶ\u0003ĨĂĐƚŽƌƐ\u0003ĐĂŶ\u0003ďĞ\u0003ĚĞĐŽŶǀŽůƵƚĞĚ͕\u0003ŝƚ\u0003\nis useful to ĮƌƐƚ\u0003understand the biophysical ďŝŶĚŝŶŐ\u0003ĐŚĂƌĂĐƚĞƌŝƐƟĐƐ\u0003ŽĨ\u0003'Z,>Ϯ\u0003ĂůŽŶĞ͘\u0003Current ChIP-\nseq and epigenomic approaches are limited in their ability to resolve detailed binding \nĚĞƚĞƌŵŝŶĂŶƚƐ͕\u0003 ŇĂŶŬŝŶŐ-ƐĞƋƵĞŶĐĞ\u0003 ĞīĞĐƚƐ͕\u0003 ĂŶĚ\u0003 ŝŶƚĞƌ-site dependencies within GRHL2-bound \nregions. Because ChIP-seq measures protein occupancy in the presence of cofactors and \nĐŚƌŽŵĂƟŶ\u0003ƐƚƌƵĐƚƵƌĞ͕\u0003ŝƚ\u0003ĐĂŶŶŽƚ\u0003ĚŝƐƟŶŐƵŝƐŚ\u0003ĚŝƌĞĐƚ\u0003ƐĞƋƵĞŶĐĞ-driven binding from cofactor-mediated \nŝ Ŷ ƚ Ğ ƌ Ă Đ Ɵ Ž Ŷ Ɛ \u0003Ž ƌ \u0003Ɖ Ă ƌ Ɛ Ğ \u0003ƚ Ś Ğ \u0003Đ Ž Ŷ ƚ ƌ ŝ ď Ƶ Ɵ Ž Ŷ \u0003Ž Ĩ \u0003ů Ž Đ Ă ů \u0003Ɛ Ğ Ƌ Ƶ Ğ Ŷ Đ Ğ \u0003ǀ Ă ƌ ŝ Ă Ɵ Ž Ŷ\u0003on binding ƐƉĞĐŝĮĐŝƚLJ\u0003 ĂŶĚ\u0003\nĂĸŶŝƚLJ. \u0018ĞĮŶŝŶŐ the GRHL2 binding preferences in detail will allow for the design of small \nmolecules that directly target GRHL2-\u0018E\u0004\u0003ŝŶƚĞƌĂĐƟŽŶƐ\u0003Ăƚ\u0003ƉƌĞĐŝƐĞ\u0003ƐŝƚĞƐ ƚŽ\u0003ŵŽĚŝĨLJ\u0003ƚŚĞ\u0003ĞdžƉƌĞƐƐŝŽŶ\u0003\nof oncogenic factors.  \n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted April 18, 2026. ; https://doi.org/10.64898/2026.04.16.719077doi: bioRxiv preprint \n\n5 \n \n In this study, GRHL2- \u0018E\u0004\u0003ŝŶƚĞƌĂĐƟŽŶƐ\u0003ĂƌĞ\u0003ĞdžĂŵŝŶĞĚ\u0003using genomic binding sites, as well \nĂƐ\u0003Ăůů\u0003ƉĞƌŵƵƚĂƟŽŶƐ\u0003ŽĨ\u0003ŵŽŶŽŵĞƌŝĐ\u0003ĂŶĚ\u0003ĚŝŵĞƌŝĐ-spaced sites. Binding events were assessed using \n^ƉĞĐŝĮĐŝƚLJ\u0003ĂŶĚ\u0003\u0004ĸŶŝƚLJ\u0003ĨŽƌ\u0003WƌŽƚĞŝŶ\u0003;^E\u0004WͿ-binding DNA microarray technology (29, 33, 34). The \nSNAP array is a custom genomic DNA array of up to 1 million features that is ďŝŽŝŶĨŽƌŵĂƟĐĂůůLJ\u0003\ndesigned using overlapping peaks from four GRHL2 ChIP-seq data sets (12, 15, 30, 35), ƵůƟŵĂƚĞůLJ \ndisplaying 48-mer probes ƚŚĂƚ\u0003ƟůĞ\u0003across 6,151 regions at 6-bp spacing using 772,732 features of \nthe array ŝ Ŷ \u0003ƚ Ś Ğ \u0003Đ Ƶ ƌ ƌ Ğ Ŷ ƚ \u0003Đ Ž Ŷ Į Ő Ƶ ƌ Ă Ɵ Ž Ŷ. This technology enables ĞdžĂŵŝŶĂƟŽŶ\u0003 ŽĨ GRHL2 binding \nacross a wide range of genomic and control sequences in greater detail. Protein-\u0018E\u0004\u0003ŝŶƚĞƌĂĐƟŽŶƐ\u0003\nare studied ǁŝƚŚŽƵƚ\u0003ƚŚĞ\u0003ŝŶŇƵĞŶĐĞ\u0003ŽĨ\u0003ŽƚŚĞƌ\u0003ĨĂĐƚŽƌƐ\u0003ƐƵĐŚ\u0003ĂƐ\u0003ĐŽĂĐƟǀĂƚŽƌƐ\u0003Žƌ\u0003ĐŽ-binding proteins that \nmay convolute the results. \u0011 LJ \u0003ŝ Ŷ ƚ Ğ Ő ƌ Ă Ɵ Ŷ Ő \u0003^ E \u0004 W \u0003Ă ƌ ƌ Ă LJ \u0003ď ŝ Ŷ Ě ŝ Ŷ Ő \u0003ŵ Ğ Ă Ɛ Ƶ ƌ Ğ ŵ Ğ Ŷ ƚ Ɛ \u0003ǁ ŝ ƚ Ś \u0003\u0012 Ś / W-seq \noccupancy, direct͕\u0003ŵŽƟĨ-directed GRHL2 binding ĐĂŶ\u0003ďĞ\u0003ĚŝīĞƌĞŶƟĂƚĞĚ\u0003from indirect recruitment \nto DNA to resolve the sequence ĐŚĂƌĂĐƚĞƌŝƐƟĐƐ ƚŚĂƚ\u0003ŝŶŇƵĞŶĐĞ\u0003'Z,>Ϯ\u0003binding. In this study, we \nĚĞĮŶĞ\u0003ƐƉĞĐŝĮĐŝƚLJ ƌƵůĞƐ\u0003ŽĨ\u0003'Z,>Ϯ\u0003ŵŽƟĨ\u0003ƌĞĐŽŐŶŝƟŽŶ\u0003ĂŶĚ ĞdžĂŵŝŶĞ\u0003ŚŽǁ\u0003ůŽĐĂů\u0003ƐĞƋƵĞŶĐĞ\u0003ĞŶǀŝƌŽŶŵĞŶƚ\u0003\nĂŶĚ\u0003ŵŽƟĨ\u0003ŽƌŐĂŶŝnjĂƟŽŶ\u0003ĐŽŶƚƌŝďƵƚĞ\u0003ƚŽ\u0003'Z,>Ϯ\u0003binding to genomic DNA. \n\u0003\nD\u0004d\u001cZ/\u0004>^\u0003\u0004E\u0018\u0003D\u001cd,K\u0018^\u0003\nGRHL2 SNAP Array Design  \n  The SNAP array (Proteovista LLC), a custom genomic DNA array synthesized by Agilent \nTechnologies͕\u0003 ǁĂƐ\u0003 ďŝŽŝŶĨŽƌŵĂƟĐĂůůLJ\u0003 ĚĞƐŝŐŶĞĚ\u0003 ƚŽ\u0003 ƟůĞ\u0003 ĂĐƌŽƐƐ\u0003 'Z,>Ϯ\u0003 ďŝŶĚŝŶŐ\u0003 ƐŝƚĞƐ\u0003based on \noverlapping peaks from four GRHL2 ChIP-seq data sets (12, 15, 30, 35). Overlapping peaks were \nŵĞƌŐĞĚ\u0003 ŝŶƚŽ\u0003 ĐŽŶƟŐƵŽƵƐ\u0003 ƌĞŐŝŽŶƐ͕\u0003 ĂŶĚ\u0003 ŐĞŶŽŵŝĐ\u0003 ŝŶƚĞƌǀĂůƐ\u0003 ĞdžĐĞĞĚŝŶŐ\u00031 kb ǁĞƌĞ\u0003 ĞdžĐůƵĚĞĚ͘\u0003Of the \nϴ͕ϵϯϭ\u0003ĐŽŵďŝŶĞĚ\u0003ŐĞŶŽŵŝĐ\u0003ƌĞŐŝŽŶƐ\u0003ŝĚĞŶƟĮĞĚ that overlapped with all four ChIP-seq datasets, 6,151 \n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted April 18, 2026. ; https://doi.org/10.64898/2026.04.16.719077doi: bioRxiv preprint \n\n6 \n \nǁ Ğƌ Ğ\u0003Ƶ Ɵ ůŝnj Ğ Ě \u0003ƚ Ž\u0003Ě ĞƐ ŝŐ Ŷ \u0003\u0018E \u0004\u0003Ăƌƌ Ă LJ\u0003Ɖ ƌ Ž ď ĞƐ \u0003Ɵ ůŝŶ Ő \u0003ĂĐƌ Ž Ɛ Ɛ \u0003Ă ƚ\u0003ϲ-bp spacing. Of these, 3401 regions \nǁĞƌĞ\u0003ƉƌŝŽƌŝƟnjĞĚ\u0003ƚŽ\u0003ŝŶĐůƵĚĞ\u0003Ă\u0003ƐƵďƐĞƚ\u0003ǁŝƚŚ\u0003ŬŶŽǁŶ\u0003'Z,>Ϯ\u0003ďŝŶĚŝŶŐ\u0003ŵŽƟĨƐ ĐŽŶƐŝƐƟŶŐ\u0003ŽĨ\u0003ƚŚĞ\u0003ĐŽŶƐĞŶƐƵƐ\u0003\n5 ’ - A A C C G G T T - 3 ’  s i t e  w i t h  u p  t o  o n e  m i s m a t c h  and regions shown previously by SNAP data \n(unpublished data) to have ER ɲͬ&Ždž\u0004ϭ\u0003 ĐŽ-binding sites. The remaining 2750 regions were \nselected based on size. dŚĞ\u0003ƉƌŽďĞƐ\u0003ĐŽŶƐŝƐƚĞĚ\u0003ŽĨ\u0003ϰϴ\u0003ŶƵĐůĞŽƟĚĞƐ\u0003ĨƌŽŵ\u0003ƚŚĞ\u0003ŐĞŶŽŵŝĐ\u0003ƌĞŐŝŽŶƐ\u0003ǁŝƚŚ\u0003Ă\u0003\n12-ŶƵĐůĞŽƟĚĞ\u0003 ĐŽŶƐƚĂŶƚ\u0003 ƌĞŐŝŽŶ\u0003 ŽŶ\u0003 ƚŚĞ\u0003 ϯ͛\u0003surface-ĂƩĂched terminus ƵƐĞĚ\u0003 ĨŽƌ\u0003 ƉƌŝŵĞƌ\u0003 ĞdžƚĞŶƐŝŽŶ͘\u0003\nControl groups, comprised of 195,282 probes, ŝŶĐůƵĚĞĚ\u0003Ăůů\u0003ƉĞƌŵƵƚĂƟŽŶƐ\u0003ŽĨ\u0003ƚŚĞ\u0003ĐŽŶƐĞŶƐƵƐ\u0003ƐŝƚĞ\u0003ĂƐ\u0003\nĂ\u0003 ŵŽŶŽŵĞƌ\u0003 ĂŶĚ\u0003 ĚŝŵĞƌ͕\u0003 ƐƉĂĐŝŶŐ\u0003 ǀĂƌŝĂƟŽŶƐ\u0003 ďĞƚǁĞĞŶ\u0003 ĚŝŵĞƌ\u0003 ƐŝƚĞƐ, sites in forward and reverse \nŽƌŝĞŶƚĂƟŽŶƐ͕\u0003ĂŶĚ\u0003ďŝŶĚŝŶŐ\u0003ƐŝƚĞƐ\u0003ŝŶ\u0003ĐŽŶũƵŶĐƟŽŶ\u0003ǁŝƚŚ\u0003&Ždž\u0004ϭ and ERɲ. A subset of 53,328 probes was \nĚĞƐŝŐŶĞĚ\u0003ĂƐ\u0003Ă\u0003ƐŝŶŐůĞ\u0003ŽůŝŐŽŶƵĐůĞŽƟĚĞ\u0003ƚŚĂƚ\u0003ĐĂŶ\u0003ĨŽƌŵ\u0003Ă\u0003ŚĂŝƌƉŝŶ\u0003ŽĨ\u0003Ϯϱ\u0003ďƉ (29, 33, 34) with a three-\nŶƵĐůĞŽƟĚĞ\u0003ůŽŽƉ, while the remaining probes were designed to be primer-ĞdžƚĞŶĚĞĚ\u0003ƚŽ\u0003Ă\u0003ůĞŶŐƚŚ\u0003ŽĨ\u0003\n48 bp. \n \nArray Extension \n To prepare the SNAP probes, the array was incubated in a conical tube of preheated 7 M \nurea ŝŶ\u0003 ϭdž\u0003phosphate-ďƵīĞƌĞĚ\u0003 ƐĂůŝŶĞ\u0003 ;PBS) at 65 \noC for 30 minutes, with brief shaking at 10-\nminute intervals. The aƌƌĂLJ\u0003ǁĂƐ\u0003ƚŚĞŶ\u0003ƚƌĂŶƐĨĞƌƌĞĚ\u0003ƚŽ\u0003ƉƌĞŚĞĂƚĞĚ\u0003ϭdž\u0003W\u0011^\u0003ĂŶĚ\u0003ŝŶĐƵďĂƚĞĚ\u0003Ăƚ\u0003ϲϱ\u0003oC for \n15 minutes, with brief shaking at 7.5 minutes ͘\u0003 EĞdžƚ͕\u0003the array was incubated in non-stringent \nǁĂƐŚ\u0003 ďƵīĞƌ\u0003 ;NSWB; ϲdž\u0003Saline-Sodium Phosphate-EDTA (SSPE) н\u0003 Ϭ͘Ϭϭй\u0003 dǁĞĞŶ-20 in ultra-pure \nwater) for 15 minutes at room temperature covered froŵ\u0003ůŝŐŚƚ\u0003ďĞĨŽƌĞ\u0003ŝƚ\u0003ǁĂƐ\u0003ƌŝŶƐĞĚ\u0003ďƌŝĞŇLJ\u0003ŝŶ\u0003ŝĐĞ\u0003\nĐŽůĚ\u0003 ϭdž\u0003 &ŝŶĂů\u0003 tĂƐŚ\u0003 \u0011ƵīĞƌ\u0003(FWB; Ϭ͘ϲdž\u0003Saline Sodium Citrate (SSC)) and dried. Phusion DNA \npolymerase (New England Biolabs, M0530L) was used with a 12- ŶƵĐůĞŽƟĚĞ\u0003 ƉƌŝŵĞƌ\u0003\n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted April 18, 2026. ; https://doi.org/10.64898/2026.04.16.719077doi: bioRxiv preprint \n\n7 \n \nĐŽŵƉůĞŵĞŶƚĂƌLJ\u0003ƚŽ\u0003ƚŚĞ\u0003ĐŽŶƐƚĂŶƚ\u0003ƉƌŽďĞ\u0003ƌĞŐŝŽŶ\u0003ƚŽ\u0003ĞdžƚĞŶĚ\u0003ƚŚĞ\u0003ƐŝŶŐůĞ-stranded genomic DNA probes \nusing published procedures (36, 37), forming 60-bp double-stranded DNA probes. Proper \nĞdžƚĞŶƐŝŽŶ\u0003was assessed on a control array using ŇƵŽƌĞƐĐĞŶƚ\u0003 \u0012LJϯ-labeled dUTP (Cy3-dUTP; \nSupplementary Figure 1). \n \nProtein \u0011inĚing anĚ DeteĐƟon \n \u0004ŌĞƌ\u0003primer ĞdžƚĞŶƐŝŽŶ͕\u0003Ă\u0003ŚLJďƌŝĚŝnjĂƟŽŶ\u0003ĐŚĂŵďĞƌ (Grace BioLabs, 623507) was mounted \nŽǀĞƌ\u0003ƚŚĞ\u0003ĂƌƌĂLJ͕\u0003ŚLJĚƌĂƚĞĚ\u0003ƚǁŝĐĞ\u0003ǁŝƚŚ\u0003ƵůƚƌĂƉƵƌĞ\u0003ǁĂƚĞƌ͕\u0003ĂŶĚ\u0003ďůŽĐŬĞĚ\u0003ǁŝƚŚ\u0003Ϯ͘ϱй\u0003ŶŽŶ-fat dry milk for 1 \nŚŽƵƌ\u0003ǁŝƚŚ\u0003ƌŽƚĂƟŽŶ͘\u0003&ŽůůŽǁŝŶŐ\u0003ďůŽĐŬŝŶŐ͕\u0003ĂƌƌĂLJƐ\u0003ǁĞƌĞ\u0003ƌŝŶƐĞĚ\u0003ǁŝƚŚ\u0003'Z,>Ϯ\u0003ďŝŶĚŝŶŐ\u0003ďƵīĞƌ\u0003;WƌŽƚĞŽǀŝƐƚĂ\u0003\n>>\u0012Ϳ͕\u0003 ĂŶĚ\u0003 ƚŚĞŶ\u0003 Ϯϱ\u0003 ŶD\u0003 ƉƵƌŝĮĞĚ\u0003 'Z,>Ϯ\u0003 ƉƌŽƚĞŝŶ\u0003 ;Krigene, TP314498 Ϳ\u0003 ĚŝůƵƚĞĚ\u0003 ŝŶ\u0003 ďŝŶĚŝŶŐ\u0003 ďƵīĞƌ\u0003\nĐ Ž Ŷ ƚ Ă ŝ Ŷ ŝ Ŷ Ő \u0003Ϭ ͘ Ϯ й \u0003\u0011 ^ \u0004 \u0003Ɖ ů Ƶ Ɛ \u0003\u0004 ů Ğ dž Ă \u0003& ů Ƶ Ž ƌ \u0003ϲ ϰ ϳ-ůĂďĞůĞĚ\u0003 ĂŶƟ-DYKDDDDK (e.g. FLAG) ĂŶƟďŽĚLJ\u0003 ;\u0012Ğůů\u0003\nSignaling, 15009S) was applied for 1 hour with rotaƟŽŶ͘\u0003The protein-ĂŶƟďŽĚLJ\u0003ƐŽůƵƟŽŶ\u0003ǁĂƐ\u0003then \nremoved, ĂŶĚ\u0003 ƚŚĞ\u0003 ĂƌƌĂLJ\u0003 ĐŚĂŵďĞƌ\u0003 ƌŝŶƐĞĚ\u0003 ǁŝƚŚ\u0003 ďŝŶĚŝŶŐ\u0003 ďƵīĞƌ ͘\u0003 dŚĞ\u0003 ĂƌƌĂLJ\u0003 ĐŚĂŵďĞƌ\u0003 ǁĂƐ\u0003 ƚŚĞŶ\u0003\nƌĞŵŽǀĞĚ͕\u0003 ĂŶĚ\u0003 ƚŚĞ\u0003 ĞŶƟƌĞ\u0003 ĂƌƌĂLJ\u0003 ďƌŝĞŇLJ\u0003 ƌŝŶƐĞĚ\u0003 ŝŶ\u0003 Ă\u0003 ĐŽŶŝĐĂů\u0003 ƚƵďĞ\u0003 ŽĨ\u0003ŝ Đ Ğ \u0003Đ Ž ů Ě \u0003ϭ dž \u0003& t \u0011 ͕ \u0003Ě ƌ ŝ Ğ Ě ͕ \u0003Ă Ŷ Ě \u0003\nƐĐĂŶŶĞĚ\u0003ƵƐŝŶŐ\u0003Ă\u0003'ĞŶĞWŝdž\u0003ϰϬϬϬ\u0011\u0003ƐĐĂŶŶĞƌ\u0003;DŽůĞĐƵůĂƌ\u0003\u0018ĞǀŝĐĞƐͿ\u0003ĂŶĚ\u0003'ĞŶĞWŝdž\u0003WƌŽ\u0003ϳ\u0003ŝŵĂŐĞ\u0003ĂĐƋƵŝƐŝƟŽŶ\u0003\nƐŽŌǁĂƌĞ\u0003;DŽůĞĐƵůĂƌ\u0003\u0018ĞǀŝĐĞƐͿ͘ \n \nArray Data NorŵaůinjaƟon  \nZĂǁ\u0003\u0004ŐŝůĞŶƚ\u0003ŵŝĐƌŽĂƌƌĂLJ\u0003ŝŵĂŐĞƐ\u0003ǁĞƌĞ\u0003ĞdžƚƌĂĐƚĞĚ\u0003ǁŝƚŚ\u0003\u0004ŐŝůĞŶƚ\u0003&ĞĂƚƵƌĞ\u0003\u001cdžƚƌĂĐƟŽŶ\u0003^ŽŌǁĂƌĞ\u0003\n(version 12.1)͘\u0003'Z,>Ϯ\u0003^E\u0004W\u0003ŵŝĐƌŽĂƌƌĂLJ\u0003ĞdžƉĞƌŝŵĞŶƚƐ\u0003ǁĞƌĞ\u0003ƉĞƌĨŽƌŵĞĚ\u0003ŝŶ\u0003ƚƌŝƉůŝĐĂƚĞ\u0003ĂŶĚ\u0003ŶŽƌŵĂůŝnjĞĚ͘\u0003\nZĂǁ\u0003ĞdžƚƌĂĐƚĞĚ\u0003ĚĂƚĂ\u0003ǁĞƌĞ\u0003ŐůŽďĂůůLJ\u0003ŶŽƌŵĂůŝnjĞĚ\u0003ƚŽ\u0003ĂĐĐŽƵŶƚ\u0003ĨŽƌ\u0003ŽǀĞƌĂůů\u0003ĚŝīĞƌĞŶĐĞƐ\u0003ŝŶ\u0003ĂďƐŽůƵƚĞ\u0003ĚĂƚĂ\u0003\nƌĂŶŐĞƐ\u0003ďĞƚǁĞĞŶ\u0003ƌĞƉůŝĐĂƚĞƐ\u0003ĂŶĚ\u0003ƚŚĞŶ\u0003ƋƵĂŶƟůĞ\u0003ŶŽƌŵĂůŝnjĞĚ\u0003ƚŽ\u0003ĂĐĐŽƵŶƚ\u0003ĨŽƌ\u0003ĚŝīĞƌĞŶĐes in intensity \n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted April 18, 2026. ; https://doi.org/10.64898/2026.04.16.719077doi: bioRxiv preprint \n\n8 \n \nĚŝƐƚƌŝďƵƟŽŶƐ\u0003(29)͘\u0003&ŝŶĂůůLJ͕\u0003ƚŚĞ\u0003ŵĞĚŝĂŶ\u0003ŝŶƚĞŶƐŝƚLJ\u0003ŽĨ\u0003ƚŚĞ\u0003ƚŚƌĞĞ\u0003ƋƵĂŶƟůĞ\u0003ŶŽƌŵĂůŝnjĞĚ\u0003ǀĂůƵĞƐ\u0003ĨŽƌ\u0003ĞĂĐŚ\u0003\nƉƌŽďĞ\u0003ǁĂƐ\u0003ƚĂŬĞŶ\u0003ĂƐ\u0003ŝƚƐ\u0003ĮŶĂů\u0003ŶŽƌŵĂůŝnjĞĚ\u0003ŝŶƚĞŶƐŝƚLJ͘\u0003dŚĞƐĞ\u0003ĮŶĂů\u0003ŶŽƌŵĂůŝnjĞĚ\u0003ŝŶƚĞŶƐŝƟĞƐ\u0003ǁĞƌĞ\u0003ƵƐĞĚ\u0003ŝŶ\u0003\nall subsequent analyses. \n \nData Analysis  \n Dimer binding model: &Žƌ\u0003 ĨĞĂƚƵƌĞƐ\u0003 ĐŽŶƚĂŝŶŝŶŐ\u0003 ƚǁŽ\u0003 'Z,>Ϯ\u0003 ďŝŶĚŝŶŐ\u0003 ŵŽƟĨƐ\u0003 ƐĞƉĂƌĂƚĞĚ\u0003 ďLJ\u0003\nĚĞĮŶĞĚ\u0003ŶƵŵďĞƌƐ\u0003ŽĨ\u0003ďĂƐĞ\u0003ƉĂŝƌƐ͕\u0003ŇƵŽƌĞƐĐĞŶĐĞ\u0003ĂƐ\u0003Ă\u0003ĨƵŶĐƟŽŶ\u0003ŽĨ\u0003ŵŽƟĨ\u0003ƐƉĂĐŝŶŐ\u0003ǁĂƐ\u0003Įƚ\u0003ǁŝƚŚ\u0003Ă\u0003\u0018E\u0004\u0003\nƚŽƌƐŝŽŶĂů\u0003ĞůĂƐƟĐŝƚLJ-based helical phasing model (38-41). In this model, the periodic dependence \nŽĨ\u0003ŇƵŽƌĞƐĐĞŶĐĞ\u0003ŽŶ\u0003ŵŽƟĨ\u0003ƐƉĂĐŝŶŐ\u0003ŝƐ\u0003ŐŽǀĞƌŶĞĚ\u0003ďLJ\u0003ƚŚĞ\u0003\u0018E\u0004\u0003ŚĞůŝĐĂů\u0003ƌĞƉĞĂƚ\u0003;௥) and is broadened by \nƚŚĞƌŵĂů\u0003 ƚŽƌƐŝŽŶĂů\u0003 ŇƵĐƚƵĂƟŽŶƐ, where ௠௜௡ Ɛ Ğ ƚ Ɛ \u0003ƚ Ś Ğ \u0003Ɖ Ś Ă Ɛ Ğ \u0003Ž ī Ɛ Ğ ƚ \u0003Ž Ĩ \u0003ƚ Ś Ğ \u0003Į ƌ Ɛ ƚ \u0003ŵ ŝ Ŷ ŝ ŵ Ƶ ŵ. The \nƉƌĞĚŝĐƚĞĚ\u0003ŇƵŽƌĞƐĐĞŶĐĞ\u0003Ăƚ\u0003ƐƉĂĐŝŶŐ\u0003Ɛ\u0003ŝƐ͗ \nF(s) = 1\n௠௜௡\n൬െ(௠௜௡+௥)ଶ\n2்ௐ(ݏ)ଶ ൰ே\n௡ୀିே\n \ndŚĞ\u0003ƐƵŵŵĂƟŽŶ\u0003ǁĂƐ\u0003ĞǀĂůƵĂƚĞĚ\u0003ĨƌŽŵ\u0003݊=ܰ ݋ݐ ܰ ,with ܰ96,  where ௥ is the helical repeat  \nƚŚĂƚ\u0003ŝƐ\u0003ĚĞĮŶĞĚ\u0003ĂƐ\u0003bp per turn, ௠௜௡ ŝƐ\u0003Ă\u0003ƐĐĂůŝŶŐ\u0003ƉĂƌĂŵĞƚĞƌ\u0003ĐŽƌƌĞƐƉŽŶĚŝŶŐ\u0003ƚŽ\u0003ƚŚĞ\u0003ĞīĞĐƟǀĞ\u0003ďŝŶĚŝŶŐ\u0003\nstrength at a phasing minimum, and ்ௐ ŝ Ɛ \u0003ƚ Ś Ğ \u0003Ɛ ƚ Ă Ŷ Ě Ă ƌ Ě \u0003Ě Ğ ǀ ŝ Ă Ɵ Ž Ŷ \u0003Ž Ĩ \u0003ƚ Ś Ğ \u0003ƌ Ğ ů Ă Ɵ ǀ Ğ \u0003ƚ ǁ ŝ Ɛ ƚ \u0003Ă Ŷ Ő ů Ğ \u0003\nďĞƚǁĞĞŶ\u0003ŵŽƟĨƐ that ŝŶĐƌĞĂƐĞƐ\u0003ǁŝƚŚ\u0003ƐƉĂĐŝŶŐ\u0003ĚƵĞ\u0003ƚŽ\u0003\u0018E\u0004\u0003ƚŽƌƐŝŽŶĂů\u0003ĞůĂƐƟĐŝƚLJ͘ \n Thermal broadening was modeled using: \n்ௐ(ݏ=)௕௣\nଶ \n௕௣=௥\n2ߨܶ݇\n௔௣௣\n \n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted April 18, 2026. ; https://doi.org/10.64898/2026.04.16.719077doi: bioRxiv preprint \n\n9 \n \nwhere ௔௣௣ is the apparent torsional modulus, κ is the rise per base pair, ݇is Boltzmann’s \nconstant, and ܶis temperature. Model parameters (௥,௔௣௣,௠௜௡,௠௜௡Ϳ\u0003ǁĞƌĞ\u0003Įƚ\u0003with nonlinear \nleast-ƐƋƵĂƌĞƐ\u0003ŵŝŶŝŵŝnjĂƟŽŶ\u0003ƚŽ\u0003ƚŚĞ\u0003ĨŽƌǁĂƌĚ-forward data ƵƐŝŶŐ\u0003ƚŚĞ\u0003ŶůŵŝŶď\u0003ĨƵŶĐƟŽŶ\u0003ŝŶ\u0003Z ƐŽŌǁĂƌĞ\u0003\n(v4.4.3). \nEƵĐůĞŽƟĚĞͬDŽƟĨ\u0003\u0018ŝƐĐƌŝŵŝŶĂƟŽŶ\u0003\u0004ŶĂůLJƐŝƐ͗ &Žƌ\u0003ĞĂĐŚ\u0003ƉŽƐŝƟŽŶ\u0003ŝŶ\u0003ƚŚĞ\u0003ůĞŌ\u0003ŇĂŶŬ\u0003;-5 to -1) and \nƌŝŐŚƚ\u0003ŇĂŶŬ\u0003;нϭ\u0003ƚŽ\u0003нϱͿ\u0003ŽĨ\u0003the consensus GRHL2 binding site, the base frequencies for the highest \nŝŶƚĞŶƐŝƚLJ\u0003ƋƵĂƌƟůĞ\u0003ĂŶĚ\u0003ůŽǁĞƐƚ\u0003ŝŶƚĞŶƐŝƚLJ\u0003ƋƵĂƌƟůĞ\u0003ǁĞƌĞ\u0003ĐĂůĐƵůĂƚĞĚ͘\u0003dŚĞ\u0003ŶƵĐůĞŽƟĚĞ\u0003ĞŶƌŝĐŚŵĞŶƚ\u0003ƌĂƟŽƐ\u0003\nĂƚ\u0003ĞĂĐŚ\u0003ŇĂŶŬŝŶŐ\u0003ƉŽƐŝƟŽŶ͕\u0003ĂƐ\u0003ĚĞĮŶĞĚ\u0003ĂƐ\u0003ƚŚĞ\u0003ƉƌŽƉŽƌƟŽŶ\u0003ŽĨ\u0003ŚŝŐŚĞƐƚ\u0003ƚŽ\u0003ůŽǁĞƐƚ\u0003ƋƵĂƌƟůĞ\u0003ĨƌĞƋƵĞŶĐŝĞƐ͕\u0003\nwere then determine Ě͘\u0003 ^ƚĂƟƐƟĐĂů\u0003 ƐŝŐŶŝĮĐĂŶĐĞ\u0003 ĂĐƌŽƐƐ\u0003 ƚŚĞ\u0003 ĨŽƵƌ\u0003 ŶƵĐůĞŽƟĚĞƐ\u0003 Ăƚ\u0003 ĞĂĐŚ\u0003 ƉŽƐŝƟŽŶ\u0003 ;Ɖ-\nvalue < 0.05) was determined with th e chi-squared test (p-value < 0.05) Ăƚ\u0003 ĞĂĐŚ\u0003 ŇĂŶŬŝŶŐ\u0003\nŶƵĐůĞŽƟĚĞ\u0003 ƉŽƐŝƟŽŶ, wherein ƚŚĞ\u0003 ĞŶƌŝĐŚŵĞŶƚ\u0003 ƌĂƟŽ\u0003of Ğ Ă Đ Ś \u0003Ŷ Ƶ Đ ů Ğ Ž Ɵ Ě Ğ \u0003Ă ƚ \u0003Ğ Ă Đ Ś \u0003Ɖ Ž Ɛ ŝ Ɵ Ž Ŷ \u0003ǁ Ă Ɛ \u0003\ncompared to its overall background ĞŶƌŝĐŚŵĞŶƚ\u0003ƌĂƟŽ\u0003across all genomic regions displayed on the \nSNAP array. For this analysis, only genomic probes with a perfect GRHL2 consensus site (5’-\nAACCGGTT-ϯ͛Ϳ\u0003ǁĞƌĞ\u0003ƵƐĞĚ͘\u0003/Ŷ\u0003ĂĚĚŝƟŽŶ͕\u0003ƉƌŽďĞƐ\u0003ǁŝƚŚ\u0003Ă\u0003ƐĞĐŽŶĚ\u0003ƉĞƌĨĞĐƚ\u0003ŵŽƟĨ\u0003or one mismatch GRHL2 \nŵŽƟĨ\u0003 ǁĞƌĞ\u0003 ĞdžĐůƵĚĞĚ\u0003 ƚŽ\u0003 ĂǀŽŝĚ\u0003 ĚŝŵĞƌŝĐ\u0003 ďŝŶĚŝŶŐ\u0003 ƐŝƚĞ\u0003 ĞīĞĐƚƐ͘\u0003 &ŝŶĂůůLJ͕\u0003 ƉƌŽďĞƐ\u0003 ǁŚĞƌĞ\u0003 ƚŚĞ\u0003 'Z,>Ϯ\u0003\nconsensus site was ǁŝƚŚŝŶ\u0003ϭϬ\u0003ďƉ\u0003ŽĨ\u0003ƚŚĞ\u0003ƉƌŝŵĞƌ\u0003ƐŝƚĞ\u0003Žƌ\u0003ĞŶĚ\u0003ŽĨ\u0003ƚŚĞ\u0003ƉƌŽďĞ\u0003ǁĞƌĞ\u0003ĞdžĐůƵĚĞĚ\u0003ƚŽ\u0003ĂǀŽŝĚ\u0003\nƚŚĞ\u0003ĞīĞĐƚƐ\u0003ŽĨ\u0003ĐŽŶƐƚĂŶƚ\u0003ƐĞƋƵĞŶĐĞƐ\u0003Žƌ\u0003ĚƵƉůĞdž\u0003ƚĞƌŵŝŶƵƐ\u0003ŝŶƐƚĂďŝůŝƚLJ͕\u0003ƌĞƐƉĞĐƟǀĞůLJ͘\u0003 \n^ĞƋƵĞŶĐĞ\u0003^ƉĞĐŝĮĐŝƚLJ\u0003>ĂŶĚƐĐĂƉĞ\u0003;SSL): SSLs were prepared as described (34, 42). For the \nŇĂƩĞŶĞĚ\u0003Ϯ\u0018\u0003^^>͕\u0003Ă\u0003WLJƚŚŽŶ\u0003ƐĐƌŝƉƚ\u0003ǁĂƐ\u0003ǁƌŝƩĞŶ\u0003ƚŽ\u0003ĞdžƚƌĂĐƚ\u0003ƚŚĞ\u0003ƐĞƋƵĞŶĐĞƐ͕\u0003dž- and y-coordinates, and \nthe ŝŶƚĞŶƐŝƟĞƐ\u0003of the one mismatch ring from the tab- ĚĞůŝŵŝƚĞĚ\u0003ƚĞdžƚ\u0003ĮůĞ\u0003ƵŶĚĞƌůLJŝŶŐ\u0003ƚŚĞ\u0003ϯ\u0018\u0003^^>. \nThis data was then visualized in a two- ĚŝŵĞŶƐŝŽŶĂů\u0003ĨŽƌŵĂƚ\u0003ǁŝƚŚ\u0003ůŝŶĞƐ\u0003ƐĞƉĂƌĂƟŶŐ\u0003ƚŚĞ\u0003ƚƌĂŶƐŝƟŽŶƐ\u0003\nďĞƚǁĞĞŶ\u0003ŵŝƐŵĂƚĐŚ\u0003ƉŽƐŝƟŽŶƐ. \n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted April 18, 2026. ; https://doi.org/10.64898/2026.04.16.719077doi: bioRxiv preprint \n\n10 \n \n\u0018ŝīĞƌĞŶƟĂů\u0003 \u0011ŝŶĚŝŶŐ\u0003 ŽĨ\u0003 \u001cZɲ\u0003 ĂŶĚ\u0003 &Ždž\u0004ϭ\u0003 ƚŽ\u0003 ^E\u0004W\u0003 ZĞŐŝŽŶƐ͗ The 6,151 SNAP regions were \ndivided into two types: direct and indirect. dŚŝƐ\u0003ǁĂƐ\u0003ĚĞƚĞƌŵŝŶĞĚ\u0003ďLJ\u0003ƚŚĞ\u0003ƉƌŽďĞ\u0003ǁŝƚŚ\u0003ƚŚĞ\u0003ŵĂdžŝŵĂl \nnormalized SNAP intensity in each region. If the normalized intensity was greater than 3.0, then \nŝƚ\u0003ǁĂƐ\u0003ĐŽŶƐŝĚĞƌĞĚ\u0003Ă\u0003'Z,>Ϯ\u0003ĚŝƌĞĐƚ\u0003ďŝŶĚŝŶŐ\u0003ƌĞŐŝŽŶ͖\u0003ŽƚŚĞƌǁŝƐĞ͕\u0003ŝƚ\u0003ǁĂƐ\u0003ĐůĂƐƐŝĮĞĚ\u0003ĂƐ\u0003ĂŶ\u0003ŝŶĚŝƌĞĐƚ\u0003'Z,>Ϯ\u0003\nbinding region. \u0018ŝīĞƌĞŶƟĂů\u0003ďinding between the direct and indirect classes of SNAP regions using \northogonal methods were an alyzed independently for ER ɲ\u0003 ĂŶĚ\u0003 &Ždž\u0004ϭ͘ dŚĞ\u0003 ĮƌƐƚ\u0003 ĚŝīĞƌĞŶƟĂů\u0003\nanalysis was based on the number of unique regions in each class that overlapped with a ChIP-\nseq peak. dŚĞ\u0003ƐĞĐŽŶĚ\u0003ĚŝīĞƌĞŶƟĂů\u0003ĂŶĂůLJƐŝƐ\u0003ǁĂƐ\u0003ďĂƐĞĚ\u0003ŽŶ\u0003ǁŚĞƚŚĞƌ\u0003ĞĂĐŚ\u0003^E\u0004W\u0003ƌĞŐŝŽŶ\u0003ĐŽŶƚĂŝŶĞĚ\u0003at \nleast one ŐĞŶŽŵŝĐ\u0003ƐĞƋƵĞŶĐĞ\u0003ĨŽƌ\u0003ƚŚĞ\u0003ƚƌĂŶƐĐƌŝƉƟŽŶ\u0003ĨĂĐƚŽƌ ͘\u0003The generalized half site 5’-RGGTCA-3’ \n(R = A or G) was used for ERɲ, and the consensus sequence 5’-AWTRTTKRYT-3' (W = A or T; K = G \nŽƌ\u0003d͖\u0003z\u0003с\u0003\u0012\u0003Žƌ\u0003dͿ\u0003ǁĂƐ\u0003ƵƐĞĚ\u0003ĨŽƌ\u0003&Ždž\u0004ϭ͘ The chi-squared test for independence with a p-value < 0.05 \nwas used for each protein and each method.\u0003\n \nZ\u001c^h>d^\u0003\nDesign anĚ ǀaliĚaƟon oĨ a ŚigŚ-resolƵƟon genoŵiĐ SNAP array Ĩor GRHL2 DNA binding sites \nTowards ƚŚĞ\u0003 ƐƵďƐĞƋƵĞŶƚ\u0003 ŝŶƚĞƌƌŽŐĂƟŽŶ\u0003 ŽĨ\u0003 'Z,>Ϯ-DNA binding ƐƉĞĐŝĮĐŝƚLJ, the domain \nŽƌŐĂŶŝnjĂƟŽŶ\u0003ŽĨ\u0003ŵĂŵŵĂůŝĂŶ\u0003'Z,>Ϯ\u0003;ŝƐŽĨŽƌŵ\u0003ϭ͖\u0003EWͺϬϳϵϭϵϭ͘ϮͿ\u0003ǁĂƐ\u0003ĞdžĂŵŝŶĞĚ\u0003(Figure 1). GRHL2 is \nĂ\u0003ƚƌĂŶƐĐƌŝƉƟŽŶ\u0003ĨĂĐƚŽƌ\u0003ŽĨ\u0003ϲϮϱ\u0003ĂŵŝŶŽ\u0003ĂĐŝĚƐ\u0003;ĂĂͿ\u0003ĐŽŵƉŽƐĞĚ\u0003ŽĨ\u0003ĂŶ\u0003E-ƚĞƌŵŝŶĂů\u0003ƚƌĂŶƐĂĐƟǀĂƟŽŶ\u0003ĚŽŵĂŝŶ\u0003\n(aa 1–133) and a central DNA-binding domain  (DBD; aa 247–484) that belongs to the \n'ƌĂŝŶLJŚĞĂĚͬ\u0012\u0012\u0004\u0004d\u0003\u0011Ždž-Binding Protein 2 (CP2) family of winged- ŚĞůŝdž\u0003ƚƌĂŶƐĐƌŝƉƟŽŶ\u0003ĨĂĐƚŽƌƐ\u0003(10, \n43). Within the DBD, a conserved core segment (aa 416–ϰϮϰͿ\u0003ĐŽƌƌĞƐƉŽŶĚƐ\u0003ƚŽ\u0003ƚŚĞ\u0003ŚĞůŝdž-turn-ŚĞůŝdž\u0003\nelement that forms some base- ƐƉĞĐŝĮĐ\u0003ĐŽŶƚĂĐƚƐ\u0003ǁŝƚŚ\u0003ƚŚĞ\u0003'Z,>\u0003ĐŽŶƐĞŶƐƵƐ\u0003ŵŽƟĨ\u0003;ϱ഻-AACCGGTT-\n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted April 18, 2026. ; https://doi.org/10.64898/2026.04.16.719077doi: bioRxiv preprint \n\n11 \n \nϯ഻Ϳ͕\u0003ĂƐ\u0003ƌĞƉŽƌƚĞĚ\u0003ŝŶ\u0003ƉƌŝŽƌ\u0003ƐƚƌƵĐƚƵƌĂů\u0003ƐƚƵĚŝĞƐ (43) . At the C-ƚĞƌŵŝŶƵƐ͕\u0003'Z,>Ϯ\u0003ĐŽŶƚĂŝŶƐ\u0003Ă\u0003ĚŝŵĞƌŝnjĂƟŽŶ\u0003\ndomain (aa 494–ϲϮϱͿ\u0003ƚŚĂƚ\u0003ŵĞĚŝĂƚĞƐ\u0003ƐƚĂďůĞ\u0003ŚŽŵŽĚŝŵĞƌ\u0003ĨŽƌŵĂƟŽŶ͘\u0003 \n dŽ\u0003 ƐLJƐƚĞŵĂƟĐĂůůLJ\u0003 ĚĞĮŶĞ\u0003 ƚŚĞ\u0003 ƐĞƋƵĞŶĐĞ\u0003 ĚĞƚĞƌŵŝŶĂŶƚƐ\u0003 ŽĨ\u0003 'Z,>Ϯ-\u0018E\u0004\u0003 ƌĞĐŽŐŶŝƟŽŶ\u0003 ĂĐƌŽƐƐ\u0003\ngenomic sequences, we generated a custom high-density SNAP DNA-binding microarray \ncomposed of a total of 968,014 DNA features with 772,732 unique double-stranded DNA features \nderived from GRHL2 ChIP-seq datasets (Figure 2A). Conserved genomic target regions were \nselected from four independent GRHL2 ChIP-seq datasets (12, 15, 30, 35). Although each dataset \npossessed thousands of GRHL2-associated peaks, only a subset of peaks (8,931) was shared \nacross all four ĞdžƉĞƌŝŵĞŶƚƐ\u0003ĚƵĞ\u0003ƚŽ\u0003ĐŽŶƚĞdžƚ-ƐƉĞĐŝĮĐ ;Ğ͘Ő͘\u0003ĐĞůů\u0003ƚLJƉĞ\u0003ĂŶĚ\u0003ĞdžƉĞƌŝŵĞŶƚĂů\u0003ĐŽŶĚŝƟŽŶƐͿ \noccupancy (Figure 2B). To capture shared binding regions, overlapping peaks were merged, and \nƚŚĞ\u0003 ƌĞƐƵůƟŶŐ\u0003 ŐĞŶŽŵŝĐ\u0003 ŝŶƚĞƌǀĂůƐ\u0003 ǁĞƌĞ\u0003 ƟůĞĚ\u0003 Ăƚ\u0003 ϲ-bp spacing to create sets of related probe \nƐĞƋƵĞŶĐĞƐ\u0003ƚŚĂƚ\u0003ĚŝīĞƌ\u0003ŽŶůLJ\u0003ďLJ\u0003ůŽĐĂů\u0003ƐĞƋƵĞŶĐĞ\u0003ƐŚŝŌƐ͘\u0003dŚĞ\u0003ƌĞŵĂŝŶŝŶŐ\u0003195,282 DNA features of the \narray were occupied with control probes and ot her altered sequences designed to test binding \nƐƉĞĐŝĮĐŝƚLJ, including probes where GRHL2 consensus and mismatch sites were combined at \nvarious spacings with ER ɲ ĂŶĚ\u0003&Ždž\u0004ϭ\u0003ĐŽŶƐĞŶƐƵƐ\u0003ĂŶĚ\u0003ŵŝƐŵĂƚĐŚ\u0003ƐŝƚĞƐ. Since ChIP-seq peaks span \nhundreds of base pairs and GRHL2 does not bind evenly throughout the peak, the 6- ďƉ\u0003 ƟůŝŶŐ \ndesign enabled ĮŶĞ\u0003ĚŝƐƐĞĐƟŽŶ\u0003ŽĨ\u0003ŇĂŶŬŝŶŐ\u0003ƐĞƋƵĞŶĐĞ\u0003ĐŽŶƚƌŝďƵƟŽŶƐ͕\u0003ŝŶƚĞƌ-ŵŽƟĨ\u0003ƐƉĂĐŝŶŐ͕\u0003ĂŶĚ\u0003local \nbinding preferences within individual ChIP-ƐĞƋ\u0003ůŽĐŝ͘\u0003WƵƌŝĮĞĚ\u0003'Z,>Ϯ\u0003ƉƌŽƚĞin was added to the array \nŝŶ\u0003 ƚŚĞ\u0003 ĂďƐĞŶĐĞ\u0003 ŽĨ\u0003 ĐĞůůƵůĂƌ\u0003 ĐŽĨĂĐƚŽƌƐ͕\u0003 ĂŶĚ\u0003 ďŽƵŶĚ\u0003 ĨĞĂƚƵƌĞƐ\u0003 ǁĞƌĞ\u0003 ĚĞƚĞĐƚĞĚ\u0003 ƵƐŝŶŐ\u0003 ĂŶ\u0003 ĂŶƟ-FLAG \nŇƵŽƌĞƐĐĞŶƚ\u0003 ĂŶƟďŽĚLJ for ƋƵĂŶƟƚĂƟǀĞ\u0003 ŵĞĂƐƵƌĞŵĞŶƚ\u0003 ŽĨ\u0003 ďŽƚŚ\u0003 ƐĞƋƵĞŶĐĞ\u0003 ƐƉĞĐŝĮĐŝƚLJ\u0003 ĂŶĚ\u0003 ĂĸŶŝƚLJ\u0003\nĂĐƌŽƐƐ\u0003ƚŚĞ\u0003ƟůĞĚ\u0003ŐĞŶŽŵŝĐ\u0003ƐĞƋƵĞŶĐĞƐ (Figure 2C). \n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted April 18, 2026. ; https://doi.org/10.64898/2026.04.16.719077doi: bioRxiv preprint \n\n12 \n \n ZĞƉƌĞƐĞŶƚĂƟǀĞ\u0003 ŐĞŶŽŵŝĐ\u0003 ƟůŝŶŐ\u0003 ĂĐƌŽƐƐ\u0003 Ă\u0003 'Z,>Ϯ-bound region near the Ovo-like \nƚƌĂŶƐĐƌŝƉƟŽŶĂů\u0003ƌĞƉƌĞƐƐŽƌ\u0003ϭ (OVOL1) ůŽĐƵƐ\u0003ƐŚŽǁĞĚ\u0003ŵƵůƟƉůĞ\u0003ƉƌŽďĞ\u0003ǀĂƌŝĂŶƚƐ\u0003ƐƉĂŶŶŝŶŐ\u0003ƚŚĞ\u0003\u0012Ś/W-seq \nƉĞĂŬ\u0003ƚŽ\u0003ŝĚĞŶƟĨLJ\u0003ŚŝŐŚ- and low-ĂĸŶŝƚLJ\u0003ƐƵďƌĞŐŝŽŶƐ\u0003ǁŝƚŚŝŶ\u0003ƚŚĞ\u0003ƉĞĂŬ\u0003;&ŝŐƵƌĞ\u00032D). Like GRHL2, OVOL1 \nacts as a mediator of EMT in ovarian cancer cells (13) and ĞdžŚŝďŝƚƐ\u0003Ă\u0003ĐŽŶƚĞdžƚ-dependent role in \nďƌĞĂƐƚ\u0003ĐĂŶĐĞƌ\u0003ďLJ\u0003ĂĐƟŶŐ\u0003ĂƐ\u0003ĞŝƚŚĞƌ\u0003ĂŶ\u0003ŽŶĐŽŐĞŶĞ\u0003Žƌ\u0003ƚƵŵŽƌ\u0003ƌĞƉƌĞƐƐŽƌ (44, 45). Consistent with the \nƉƌĞƐĞŶĐĞ\u0003 ŽĨ\u0003 ĨƵŶĐƟŽŶĂů\u0003 'Z,>Ϯ\u0003 ŵŽƟĨƐ͕\u0003 Ă\u0003 ƐƵďƐĞƚ\u0003 ŽĨ\u0003 ƟůĞĚ\u0003 ƉƌŽďĞƐ\u0003 ŚĂĚ\u0003ƐƚƌŽŶŐ\u0003 ďŝŶĚŝŶŐ͕\u0003 ǁŚĞƌĞĂƐ\u0003\nadjacent variants, ĚŝīĞƌŝŶŐ\u0003 ŽŶůLJ\u0003 ŝŶ\u0003 ŇĂŶŬŝŶŐ\u0003 ƐĞƋƵĞŶĐĞ\u0003 Žƌ\u0003 ƉŽƐŝƟŽŶĂů\u0003 ŽīƐĞƚ, displayed reduced \nŇƵŽƌĞƐĐĞŶĐĞ\u0003ŝŶƚĞŶƐŝƚLJ͘\u0003dŚĞƐĞ\u0003ƌĞƐƵůƚƐ\u0003ŝŶĚŝĐĂƚĞ\u0003ƚŚĂƚ\u0003'Z,>Ϯ\u0003ďŝŶĚŝŶŐ\u0003ĐĂŶ\u0003ďĞ\u0003ůŽĐĂůŝnjĞĚ\u0003ǁŝƚŚŝŶ\u0003\u0012Ś/W-seq \nƉĞĂŬƐ\u0003ĂŶĚ\u0003ŝƐ\u0003ƐĞŶƐŝƟǀĞ\u0003ƚŽ\u0003ůŽĐĂů\u0003ƐĞƋƵĞŶĐĞ\u0003ĐŽŶƚĞdžƚ͘ \n&ůƵŽƌĞƐĐĞŶĐĞ\u0003 ŝŵĂŐŝŶŐ\u0003 ŽĨ\u0003 ƚŚĞ\u0003 ĂƌƌĂLJ\u0003 ĐŽŶĮƌŵĞĚ\u0003 ĚŝƐĐƌĞƚĞ͕\u0003 ƌĞƉƌŽĚƵĐŝďůĞ\u0003 ďŝŶĚŝŶŐ\u0003 ĂĐƌŽƐƐ\u0003\nƚŚŽƵƐĂŶĚƐ\u0003 ŽĨ\u0003 ƉƌŽďĞ\u0003 ĨĞĂƚƵƌĞƐ͘\u0003 \u0011ŽƵŶĚ\u0003 ĨĞĂƚƵƌĞƐ\u0003 ǁĞƌĞ\u0003 ĚŝƐƟŶŐƵŝƐŚĞĚ\u0003 ĨƌŽŵ\u0003 ďĂĐŬŐƌŽƵŶĚ͕\u0003 ĂŶĚ the \nĚŝƐƚƌŝďƵƟŽŶ\u0003 ŽĨ\u0003 ŇƵŽƌĞƐĐĞŶĐĞ ŝŶƚĞŶƐŝƟĞƐ\u0003 ǁĞƌĞ\u0003 ĐŽŶĮƌŵĞĚ\u0003 ďLJ\u0003 ƌĞƉůŝĐĂƚĞ\u0003 ĂƌƌĂLJƐ͕\u0003 ǀĂůŝĚĂƟŶŐ\u0003 ƚŚĞ\u0003\nrobustness of the SNAP-ďĂƐĞĚ\u0003ĂĸŶŝƚLJ\u0003ŵĞĂƐƵƌĞŵĞŶƚƐ͘\u0003dŚĞƐĞ\u0003ĚĂƚĂ\u0003ĞŶĂďůĞ\u0003ƋƵĂŶƟƚĂƟǀĞ\u0003ƐĞƉĂƌĂƟŽŶ\u0003\nŽ Ĩ \u0003Ě ŝ ƌ Ğ Đ ƚ ͕ \u0003ŵ Ž Ɵ Ĩ-driven GRHL2 binding from indirect or cofactor-mediated occupancy that is \nŝŶĚŝƐƟŶŐƵŝƐŚĂďůĞ\u0003ŝŶ ChIP-seq.  \n \nGRHL2 binding Ĩaǀors tŚe canonical AA \u0012\u0012GGdd ŵoƟĨ bƵt exŚibits selecƟǀe tolerance to single-\nnƵcleoƟde ǀariaƟon and ŇanŬing-sequence context \n dŚĞ\u0003 ^E\u0004W\u0003 ĂƌƌĂLJ\u0003 ǁĂƐ\u0003 ƵƐĞĚ\u0003 ƚŽ\u0003 ĚĞĮŶĞ\u0003 ƚŚĞ\u0003 ƐĞƋƵĞŶĐĞ\u0003 ĚĞƚĞƌŵŝŶĂŶƚƐ\u0003 ƚŚĂƚ\u0003 ĚŝƐƟŶŐƵŝƐŚ\u0003 ŚŝŐŚ-\nĂĸŶŝƚLJ\u0003 'Z,>Ϯ\u0003 ďŝŶĚŝŶŐ\u0003 ĨƌŽŵ\u0003 ƉĂƌƟĂůůLJ\u0003 ƚŽůĞƌĂƚĞĚ\u0003 ŵŽƟĨ\u0003 ǀĂƌŝĂƟŽŶƐ͘\u0003 dŽ\u0003Ě Ž \u0003Ɛ Ž ͕ \u0003Ă ů ů genomic probe \nĨĞĂƚƵƌĞƐ\u0003ǁĞƌĞ\u0003ƌĂŶŬĞĚ\u0003ďLJ\u0003ŇƵŽƌĞƐĐĞŶĐĞ\u0003ŝŶƚĞŶƐŝƚLJ͕\u0003ĂŶĚ\u0003ƚŚĞ\u0003ƚŽƉ\u0003ϭ͕ϮϬϬ\u0003ŚŝŐŚĞƐƚ-ĂĸŶŝƚLJ\u0003ƐĞƋƵĞŶĐĞƐ\u0003ǁĞƌĞ\u0003\nĞdžƚƌĂĐƚĞĚ\u0003ĨŽƌ\u0003ĚĞ\u0003ŶŽǀŽ\u0003ŵŽƟĨ\u0003ĚŝƐĐŽǀĞƌLJ͘\u0003\u0004ĐƌŽƐƐ\u0003four independent bins of progressively decreasing \n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted April 18, 2026. ; https://doi.org/10.64898/2026.04.16.719077doi: bioRxiv preprint \n\n13 \n \n300 top-binding probes that total the 1,200 highest- ĂĸŶŝƚLJ\u0003ŐĞŶŽŵŝĐ\u0003ƐĞƋƵĞŶĐĞƐ, MEME analysis \n(46) Đ Ž Ŷ Ɛ ŝ Ɛ ƚ Ğ Ŷ ƚ ů LJ \u0003ƌ Ğ Đ Ž ǀ Ğ ƌ Ğ Ě \u0003ƚ Ś Ğ \u0003Đ Ă Ŷ Ž Ŷ ŝ Đ Ă ů \u0003' Z , > Ϯ \u0003Đ Ž Ŷ Ɛ Ğ Ŷ Ɛ Ƶ Ɛ \u0003ŵ Ž Ɵ Ĩ \u0003\u0004 \u0004 \u0012 \u0012 ''dd ͕\u0003 ǁŝƚŚ\u0003 ŵŝŶŝŵal \nƉŽƐŝƟŽŶĂů\u0003ǀĂƌŝĂďŝůŝƚLJ\u0003;&ŝŐƵƌĞ\u0003ϯ\u0004Ϳ͘\u0003dŚĞ\u0003ĐŽŶƐĞƌǀĂƟŽŶ\u0003ŽĨ\u0003ƚŚĞ\u0003ĐŽŶƐĞŶƐƵƐ\u0003ĂĐƌŽƐƐ\u0003ƚŚĞ\u0003ĨŽƵƌ\u0003ŝŶƚĞŶƐŝƚLJ\u0003ďŝŶƐ\u0003\nŝŶĚŝĐĂƚĞƐ\u0003ƚŚĂƚ\u0003ƐƚƌŽŶŐ\u0003ďŝŶĚŝŶŐ\u0003ŝƐ\u0003ĚŽŵŝŶĂƚĞĚ\u0003ďLJ\u0003ƌĞĐŽŐŶŝƟŽŶ\u0003ŽĨ\u0003ƚŚĞ\u0003ĐĂŶŽŶŝĐĂů\u0003ŽĐƚĂŵĞƌ\u0003ŵŽƟĨ͕\u0003ƌĂƚŚĞƌ\u0003\nƚŚĂŶ\u0003ĐŽŶƚƌŝďƵƟŽŶƐ\u0003ĨƌŽŵ ŇĂŶŬŝŶŐ\u0003Žƌ\u0003genome-ĐŽŶƚĞdžƚ ƉŽƐŝƟŽŶƐ. To further ĞdžĂŵŝŶĞ\u0003ƚŚĞ\u0003ƐƉĞĐŝĮĐŝƚLJ\u0003\nof the 772,732 genomic sequences derived from the GRHL2 ChIP-seq datasets͕\u0003ǁĞ\u0003ĞdžĂŵŝŶĞĚ\u0003ƚŚĞ\u0003\n45,ϯϴϰ\u0003ƐĞƋƵĞŶĐĞƐ\u0003ƚŚĂƚ\u0003ĚŝīĞƌĞĚ\u0003ĨƌŽŵ\u0003ƚŚĞ\u0003ĐŽŶƐĞŶƐƵƐ\u0003ďLJ\u0003Ă\u0003ƐŝŶŐůĞ\u0003ŶƵĐůĞŽƟĚĞ\u0003ƐƵďƐƟƚƵƟŽŶ\u0003ǁŝƚŚŝŶ\u0003ƚŚĞ\u0003\nŵŽƟĨ͘\u0003DŽƟĨ\u0003analysis of the top binders of these ƐŝŶŐůĞ\u0003ŵƵƚĂƟŽŶ variants revealed a strong binding \ndependeŶĐĞ\u0003ŽŶ\u0003ƉŽƐŝƟŽŶƐ\u0003ϭ-3 and 6- ϴ\u0003ŽĨ\u0003ƚŚĞ\u0003ŵŽƟĨ ͕\u0003ĂůŽŶŐ\u0003ǁŝƚŚ\u0003ƐŽŵĞ\u0003ŇĂŶŬŝŶŐ\u0003ƉƌĞĨĞƌĞŶĐĞƐ\u0003ǁŚĞŶ\u0003\nƚŚĞ\u0003ďŝŶĚŝŶŐ\u0003ǁŝŶĚŽǁ\u0003ŝƐ\u0003ĞdžƉĂŶĚĞĚ\u0003ƚŽ\u0003ϭϴ\u0003ƉŽƐŝƟŽŶƐ\u0003(Figure 3B). In a separate analysis of the ŇĂŶŬŝŶŐ\u0003\nbases ƐƵƌƌŽƵŶĚŝŶŐ\u0003ƚŚĞ\u0003ƉĞƌĨĞĐƚ\u0003ŽĐƚĂŵĞƌ\u0003ŵŽƟĨ͕\u0003ƐŝŐŶŝĮĐĂŶƚ\u0003ĞŶƌŝĐŚŵĞŶƚ\u0003ĨŽƌ\u0003ƐƉĞĐŝĮĐ\u0003ϱ഻\u0003ĂŶĚ\u0003ϯ഻\u0003ďĂƐĞƐ\u0003\nǁĂƐ\u0003 ŽďƐĞƌǀĞĚ\u0003 ǁŝƚŚ\u0003 ĂŶ\u0003 ŽƉƟŵĂů\u0003 ϱ-ďĂƐĞ\u0003 ŇĂŶŬŝŶŐ\u0003 ƐĞƋƵĞŶĐĞ\u0003 ŽĨ\u0003 ϱ͛-NNNGA-AACCGGTT-CNGAG-3’ \n;&ŝŐƵƌĞ\u0003 ϯ\u0012Ϳ͘\u0003 dŚŝƐ\u0003 ƐƵŐŐĞƐƚƐ\u0003 ƚŚĂƚ\u0003 ƉƌŽdžŝŵĂů\u0003 ƐĞƋƵĞŶĐĞ\u0003 ĞŶǀŝƌŽŶŵĞŶƚ\u0003 ŵŽĚƵůĂƚĞƐ\u0003 ĂĸŶŝƚLJ\u0003 ǁŝƚŚŽƵƚ\u0003\nĂůƚĞƌŝŶŐ\u0003ƚŚĞ\u0003ĐŽƌĞ\u0003ŵŽƟĨ\u0003ƌĞĐŽŐŶŝƟŽŶ͘ \n dŽ\u0003 ǀŝƐƵĂůŝnjĞ\u0003 ŚŽǁ\u0003 ĂĚĚŝƟŽŶĂů\u0003 ƐĞƋƵĞŶĐĞ\u0003 ƉĞƌƚƵƌďĂƟŽŶƐ\u0003 ŝŶŇƵĞŶĐĞ\u0003 ƌĞĐŽŐŶŝƟŽŶ͕\u0003 Ăn SSL of all \n772,732 sequences was constructed in which the central ring is populated with array sequences \nbearing the perfect \u0004\u0004\u0012\u0012''dd\u0003 ŵŽƟĨ͕\u0003 ĂŶĚ\u0003 ĞĂĐŚ\u0003 ƐƵĐĐĞƐƐŝǀĞ\u0003 ĐŽŶĐĞŶƚƌŝĐ\u0003 ƌŝŶŐ\u0003 ĐŽŶƚĂŝŶƐ\u0003 ƐĞƋƵĞŶĐĞƐ\u0003\nƚŚĂƚ\u0003ŝŶƚƌŽĚƵĐĞ\u0003ŽŶĞ\u0003ĂĚĚŝƟŽŶĂů\u0003ŵŝƐŵĂƚĐŚ\u0003ŝŶƚŽ\u0003ƚŚĞ\u0003ŵŽƟĨ\u0003;&ŝŐƵƌĞ\u00033D) (34, 42). Array sequences were \nŽƌŐĂŶŝnjĞĚ\u0003ĂƌŽƵŶĚ\u0003ƚŚĞ\u0003ƌŝŶŐ\u0003ĂĐĐŽƌĚŝŶŐ\u0003ƚŽ\u0003ƚŚĞ\u0003ƐƉĞĐŝĮĐ\u0003ƉŽƐŝƟŽŶ\u0003ǁŝƚŚŝŶ\u0003ƚŚĞ\u0003ŵŽƟĨ\u0003Ăƚ\u0003ǁŚŝĐŚ\u0003ƚŚĞ\u0003ƐŝŶŐůĞ\u0003\nŶƵĐůĞŽƟĚĞ\u0003 ƐƵďƐƟƚƵƟŽŶ\u0003 ŽĐĐƵƌƌĞĚ͕\u0003 ĂŶĚ\u0003 ƚŚĞŶ\u0003 ďLJ\u0003 ƚŚĞ\u0003 ƐƉĞĐŝĮĐ\u0003 ŶƵĐůĞŽƟĚĞ \u0003Ž Ĩ \u0003ƚ Ś Ğ \u0003ŵ ŝ Ɛ ŵ Ă ƚ Đ Ś. Thus, \npeaks in the mismatch rings correspond to ƐƉĞĐŝĮĐ\u0003ŵŝƐŵĂƚĐŚ-ƐĞƋƵĞŶĐĞ\u0003ĐŽŵďŝŶĂƟŽŶƐ\u0003ƚŚĂƚ\u0003ĂƌĞ\u0003\nƉƌĞĨĞƌĞŶƟĂůůLJ\u0003ďŽƵŶĚ\u0003ďLJ\u0003'Z,>2, providing ŝŶĨŽƌŵĂƟŽŶ\u0003ŽŶ\u0003ďŽƚŚ\u0003ŵŽƟĨ\u0003ŝŶƚĞƌĚĞƉĞŶĚĞŶĐŝĞƐ\u0003ĂƐ\u0003ǁĞůů\u0003\n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted April 18, 2026. ; https://doi.org/10.64898/2026.04.16.719077doi: bioRxiv preprint \n\n14 \n \nĂƐ\u0003ŇĂŶŬŝŶŐ\u0003ƐĞƋƵĞŶĐĞ\u0003ĞīĞĐƚƐ͘\u0003^ŝŶĐĞ\u0003ƚŚĞ\u0003ŇƵŽƌĞƐĐĞŶĐĞ\u0003ŝŶƚĞŶƐŝƚLJ\u0003ŝƐ\u0003ƉƌŽƉŽƌƟŽŶĂů\u0003ƚŽ ƚŚĞ\u0003ĂƐƐŽĐŝĂƟŽŶ\u0003\nconstant (Ka), ƚŚĞ\u0003 ^^>\u0003 ĂůƐŽ\u0003 ƉƌŽǀŝĚĞƐ\u0003 Ă\u0003 ƋƵĂŶƟƚĂƟǀĞ\u0003 ďŝŶĚŝŶŐ\u0003 ĞŶĞƌŐLJ\u0003 ƉƌŽĮůĞ ƌĞůĂƟŶŐ ŇƵŽƌĞƐĐĞŶĐĞ \nŝŶƚĞŶƐŝƚLJ\u0003 ƚŽ\u0003 ďŝŶĚŝŶŐ\u0003 ĂĸŶŝƚLJ (34, 42) ͘\u0003 WƌŽŐƌĞƐƐŝǀĞ\u0003 ŶƵŵďĞƌƐ\u0003 ŽĨ\u0003 ŵƵƚĂƟŽŶƐ\u0003 ;ƌŝŶŐƐ\u0003 ϭ-5) produced a \nŐƌĂĚĞĚ\u0003ĚĞĐƌĞĂƐĞ\u0003ŝŶ\u0003ŇƵŽƌĞƐĐĞŶĐĞ\u0003ŝŶƚĞŶƐŝƚLJ͕\u0003ǁŝƚŚ\u0003ƚŚĞ\u0003ƐƚĞĞƉĞƐƚ\u0003ĚĞĐůŝŶĞ\u0003ŽĐĐƵƌƌŝŶŐ\u0003ǁŚĞŶ\u0003ƐƵďƐƟƚƵƟŽŶƐ\u0003\nĂīĞĐƚĞĚ\u0003ƚŚĞ\u0003\u0012\u0003ĂŶĚ\u0003'\u0003Ăƚ\u0003ƉŽƐŝƟŽŶƐ\u0003ϯ\u0003ĂŶĚ\u0003ϲ͕\u0003ƌĞƐƉĞĐƟǀĞůLJ. In contrast, a large number of mismatches \nto the consensus sequence, which are displayed in the outer rings, produced ǀĞƌLJ\u0003ůŝƩůĞ\u0003ďŝŶĚŝŶŐ\u0003\nsignal. Low- ĐŽŵƉůĞdžŝƚLJ\u0003 ƌĞƉĞĂƚ\u0003 ĐŽŶƚĞdžƚƐ\u0003 ƐƵĐŚ\u0003 ĂƐ\u0003 \u0012\u0004- or TA-ƌ ŝ Đ Ś \u0003Ň Ă Ŷ Ŭ Ɛ \u0003Ă ů Ɛ Ž \u0003modulated binding, \nǁŚŝĐŚ\u0003 ŝƐ\u0003 ĐŽŶƐŝƐƚĞŶƚ\u0003 ǁŝƚŚ\u0003 ƐŽŵĞ\u0003 ĐŽŶƚƌŝďƵƟŽŶƐ\u0003 ĨƌŽŵ\u0003 ůŽĐĂů\u0003 ƐĞƋƵĞŶĐĞ\u0003ĞŶǀŝronment. This analysis \nƌĞǀĞĂůĞĚ\u0003ŵĂdžŝŵĂů\u0003ŝŶƚĞŶƐŝƚLJ\u0003ƌĞĚƵĐƟŽŶ\u0003ďLJ\u0003ŵƵƚĂƟŽŶƐ\u0003Ăƚ\u0003ƉŽƐŝƟŽŶƐ\u0003ϯ\u0003ĂŶĚ\u0003ϲ\u0003ĐŽŵďŝŶĞĚ\u0003ǁŝƚŚ\u0003ŵŝŶŝŵĂů\u0003\nŝŶƚĞŶƐŝƚLJ\u0003ĐŚĂŶŐĞƐ\u0003Ăƚ\u0003ƉŽƐŝƟŽŶƐ\u0003ϰ-ϱ\u0003ƚŽ\u0003ŐĞŶĞƌĂƚĞ\u0003Ă\u0003ĐĞŶƚƌĂů\u0003\u0012dždž'\u0003ƐŝƚĞ\u0003ǁŝƚŚŝŶ\u0003ƚŚĞ\u0003ĐŽŶƐĞŶƐƵƐ\u0003\u0004\u0004CCGGTT \nŵŽƟĨ͘\u0003 \n dŽ\u0003 ƌĞĮŶĞ\u0003 ƚŚĞ\u0003 ƚŽůĞƌĂƚĞĚ\u0003 ǀĂƌŝĂƟŽŶƐ\u0003 ŽĨ\u0003 ƚŚĞ\u0003 ŵŽƟĨ͕\u0003 ĨƵƌƚŚĞƌ\u0003 ĂŶĂůLJƐĞƐ\u0003Ž Ĩ \u0003ƚ Ś Ğ \u0003Ž Ŷ Ğ-mismatch \n(Figure 3E) and two- ŵŝƐŵĂƚĐŚ\u0003 ƌŝŶŐƐ\u0003 ǁĞƌĞ\u0003 ƉĞƌĨŽƌŵĞĚ͘\u0003 dŚĞ\u0003 ŶƵŵďĞƌ\u0003 ĂŶĚ\u0003 ƉŽƐŝƟŽŶ\u0003 ŽĨ\u0003 ƐĞƋƵĞŶĐĞƐ\u0003\nĂƌŽƵŶĚ\u0003ƚŚĞ\u0003^^>\u0003ƌŝŶŐ\u0003ƉƌŽǀŝĚĞ\u0003ƌĞĂĚŽƵƚ\u0003ŽŶ\u0003ďĂƐĞ\u0003ƚŽůĞƌĂŶĐĞƐ͘\u0003DƵƚĂƟŽŶƐ\u0003Ăƚ\u0003ƚŚĞ\u0003\u0012\u0003;ƉŽƐŝƟŽŶ\u0003ϯͿ\u0003ĂŶĚ\u0003'\u0003\n;ƉŽƐŝƟŽŶ\u0003ϲͿ\u0003ŽĨ\u0003ƚŚĞ\u0003ĐĞŶƚƌĂů\u0003\u0012dždž'\u0003were depleted among high-ŝŶƚĞŶƐŝƚLJ\u0003ďŝŶĚĞƌƐ\u0003ĂƐ\u0003ŽŶůLJ\u0003ϭ͘ϵй\u0003ŽĨ\u0003ƚŚĞ\u0003\nŵ Ƶ ƚ Ă Ɵ Ž Ŷ Ɛ ͕ \u0003ǁ Ś ŝ ů Ğ \u0003Ɛ Ƶ ď Ɛ Ɵ ƚ Ƶ Ɵ Ž Ŷ Ɛ \u0003Ă ƚ \u0003ƚ Ś Ğ \u0003ƚ Ğ ƌ ŵ ŝ Ŷ Ă ů \u0003Ň Ă Ŷ Ŭ ŝ Ŷ Ő \u0003\u0004 ͬ d \u0003Ɖ Ž Ɛ ŝ Ɵ Ž ŶƐ\u0003 ǁĞƌĞ\u0003more tolerated, \nŝŶĚŝĐĂƟŶŐ\u0003ƚŚĂƚ\u0003ƉŽƐŝƟŽŶƐ\u0003ϯ\u0003ĂŶĚ\u0003ϲ\u0003ŽĨ\u0003the octamer ĐŽŶƚƌŝďƵƚĞ\u0003ĚŝƐƉƌŽƉŽƌƟŽŶĂƚĞůLJ\u0003ƚŽ\u0003ďŝŶĚŝŶŐ\u0003ƐƚĂďŝůŝƚLJ͘\u0003\n/ŶƚĞƌĞƐƟŶŐůLJ͕\u0003ŵƵƚĂƟŽŶƐ\u0003Ăƚ\u0003ƚŚĞ\u0003ŝŶŶĞƌ\u0003\u0012'\u0003;ƉŽƐŝƟŽŶƐ\u0003ϰ-ϱͿ\u0003ǁĞƌĞ\u0003ĞŶƌŝĐŚĞĚ\u0003Ăƚ\u0003ϯϵ͘ϱй\u0003ĞĂĐŚ, with the C \nŵŽƐƚ\u0003ŽŌĞŶ\u0003ŵƵƚĂƚĞĚ to T (CїT: ϱϬ͘ϮйͿ\u0003ĂŶĚ A (CїA: ϰϲ͘ϯйͿ, and the G conversely was ŵŽƐƚ\u0003ŽŌĞŶ\u0003\nmutated to A (G їA: ϱϬ͘ϮйͿ\u0003 ĂŶĚ\u0003T (GїT: ϰϲ͘ϯйͿ͘\u0003 dŚĞƐĞ\u0003 ĂŶĂůLJƐĞƐ\u0003 ĚĞĮŶĞ\u0003 Ă\u0003 ŚŝŐŚ-ƌĞƐŽůƵƟŽŶ\u0003\nƉŽƐŝƟŽŶĂů\u0003 ďŝŶĚŝŶŐ\u0003 ůĂŶĚƐĐĂƉĞ\u0003 ĨŽƌ\u0003 'Z,>Ϯ-\u0018 E \u0004 \u0003ŝ Ŷ ƚ Ğ ƌ Ă Đ Ɵ Ž Ŷ Ɛ ͕ \u0003Ě Ğ ŵ Ž Ŷ Ɛ ƚ ƌ Ă Ɵ Ŷ Ő \u0003ƚ Ś Ă ƚ \u0003ƚ Ś Ğ \u0003Đ Ă Ŷ Ž Ŷ ŝ Đ Ă ů \u0003\n\u0004\u0004\u0012\u0012''dd\u0003ŵŽƟĨ\u0003ŝƐ\u0003ďŽƚŚ\u0003ŶĞĐĞƐƐĂƌLJ\u0003ĂŶĚ\u0003ƐƵĸĐŝĞŶƚ\u0003ĨŽƌ\u0003ŚŝŐŚ-ĂĸŶŝƚLJ\u0003ďŝŶĚŝŶŐ͘\u0003\u0003&ƵƌƚŚĞƌŵŽƌĞ͕\u0003ƉŽƐŝƟons \n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted April 18, 2026. ; https://doi.org/10.64898/2026.04.16.719077doi: bioRxiv preprint \n\n15 \n \n3 and 6 of the octamer and shown in the central \u0012dždž'\u0003ĐŽŶƐƟƚƵƚĞ\u0003ƚŚĞ\u0003ŵŽƐƚ\u0003ƐĞƋƵĞŶĐĞ-constrained \nƌĞŐŝŽŶ\u0003ŽĨ\u0003ƚŚĞ\u0003ƐŝƚĞ͕\u0003ĂŶĚ\u0003ŇĂŶŬŝŶŐ\u0003ďĂƐĞƐ\u0003ŵŽĚƵůĂƚĞ\u0003ďƵƚ\u0003ĚŽ\u0003ŶŽƚ\u0003ŽǀĞƌƌŝĚĞ\u0003ƚŚĞ\u0003octamer ŵŽƟĨ\u0003ƌĞĐŽŐŶŝƟŽŶ͘\u0003 \n \nExtended ŵoƟĨ analysis reǀeals tolerance to ŵisŵatcŚes across diŵeric GRHL2 binding sites \n To ĚĞƚĞƌŵŝŶĞ\u0003 ǁŚĞƚŚĞƌ\u0003 'Z,>Ϯ\u0003 ƌĞĐŽŐŶŝnjĞƐ\u0003 ƐĞƋƵĞŶĐĞ\u0003 ĨĞĂƚƵƌĞƐ\u0003 ĞdžƚĞŶĚŝŶŐ\u0003ďĞLJŽŶĚ\u0003 ƚŚĞ\u0003\ncanonical \u0004 \u0004 \u0012 \u0012 ' ' d d \u0003ŵ Ž Ɵ Ĩ ͕ \u0003ǁ Ğ \u0003Ŷ Ğ dž ƚ \u0003Ɖ Ğ ƌ Ĩ Ž ƌ ŵ Ğ Ě \u0003Ğ dž ƚ Ğ Ŷ Ě Ğ Ě-ŵ Ž Ɵ Ĩ \u0003Ě ŝ Ɛ Đ Ž ǀ Ğ ƌ LJ \u0003Ƶ Ɛ ŝ Ŷ Ő \u0003ƚ Ś Ğ \u0003ƚ Ž Ɖ \u0003ϭ ͕ Ϯ Ϭ Ϭ \u0003\nhighest-intensity genomic probes of the 772,732 genomic ƉƌŽďĞ\u0003 ƐĞƚ͕\u0003 ǁŝƚŚ\u0003 Ă\u0003 ĐŽŶƐƚƌĂŝŶĞĚ\u0003 ŵŽƟĨ\u0003\nǁŝĚƚŚ\u0003ŽĨ\u0003ϭϴ\u0003ďƉ͘\u0003dŚĞ\u0003ƌĞƐƵůƟŶŐ\u0003ŵŽƟĨ\u0003ƌĞƚĂŝŶĞĚ\u0003ĂŶ\u0003\u0004\u0004\u0012\u0012''dd\u0003consensus but also revealed evidence \nŽĨ\u0003ƚǁŽ\u0003ŵŽƟĨƐ͕\u0003ƐƵŐŐĞƐƟǀĞ\u0003ŽĨ\u0003ĚŝŵĞƌ\u0003ďŝŶĚŝŶŐ\u0003ŽŶ\u0003ƚŚĞ\u0003^E\u0004W\u0003ƉƌŽďĞƐ\u0003;&ŝŐƵƌĞ\u0003ϰ\u0004Ϳ͘ \n d Ž \u0003ŝ Ŷ ƚ Ğ ƌ Ɖ ƌ Ğ ƚ \u0003ƚ Ś Ğ Ɛ Ğ \u0003Ğ dž ƚ Ğ Ŷ Ě Ğ Ě \u0003Ɖ ƌ Ğ Ĩ Ğ ƌ Ğ Ŷ Đ Ğ Ɛ \u0003ŝ Ŷ \u0003ƚ Ś Ğ \u0003Đ Ž Ŷ ƚ Ğ dž ƚ \u0003Ž Ĩ \u0003ŵ Ž ů Ğ ĐƵůĂƌ\u0003 ƌĞĐŽŐŶŝƟŽŶ͕\u0003 ǁĞ\u0003\nĞdžĂŵŝŶĞĚ\u0003ƚŚĞ\u0003'Z,>Ϯ\u0003\u0018E\u0004-binding domain crystal structure (PDB: 5MR7) overlaid with a linear \nĚƵƉůĞdž\u0003\u0018E\u0004\u0003ŵŽĚĞů\u0003ĂůŝŐŶĞĚ\u0003ƚŽ\u0003ƚŚĞ\u0003'Z,>ϭ-\u0018E\u0004\u0003ĐŽŵƉůĞdž\u0003;W\u0018\u0011͗\u0003ϱDW&Ϳ\u0003;&ŝŐƵƌĞ\u0003ϰ\u0011Ϳ since a crystal \nstructure of GRHL2-DNA is not present in RSCB Protein Data Bank. This overlay is reasonable since \nthe DNA-binding domain of these two proteins (residues 247-484 for GRHL2 from Uniprot \nQ6ISB3; residues 250-467 for GRHL1 from Uniprot Q9NZI5 Ϳ\u0003ĞdžŚŝďŝƚ\u0003ϳϲй\u0003ŝĚĞŶƟƚLJ, 94й\u0003ƐŝŵŝůĂƌŝƚLJ, \nand a single gap of 12 residues. /Ŷ\u0003ƚŚŝƐ\u0003ĐŽŶĮŐƵƌĂƟŽŶ͕\u0003ƚŚĞ\u0003ƌĞĐŽŐŶŝƟŽŶ\u0003ŚĞůŝdž\u0003ŝƐ\u0003ƉŽƐŝƟŽŶĞĚ\u0003ŽǀĞƌ\u0003ƚŚĞ\u0003\ncentral CCGG of the octamer consensus. Residues in the wing and C-ƚĞƌŵŝŶĂů\u0003ĞdžƚĞŶƐŝŽŶ\u0003ĂƉƉƌŽĂĐŚ\u0003\nthe DNA phosphate backďŽŶĞ\u0003ĂŶĚ\u0003ĂĚũĂĐĞŶƚ\u0003ŇĂŶŬŝŶŐ\u0003ďĂƐĞƐ͕\u0003ƐƵŐŐĞƐƟŶŐ\u0003ƉŽƚĞŶƟĂů\u0003ƉƌŽƚĞŝŶ\u0003ĐŽŶƚĂĐƚƐ\u0003\nŽƵƚƐŝĚĞ\u0003ƚŚĞ\u0003ĐŽƌĞ\u0003ŵŽƟĨ͘ \n hƐŝŶŐ\u0003ĐŽŶƚƌŽů\u0003ƉƌŽďĞƐ\u0003ǁŝƚŚ\u0003Ă\u0003ŵŽƟĨ\u0003ĞŵďĞĚĚĞĚ\u0003ǁŝƚŚŝŶ\u0003Ă\u0003ĐŽŶƐƚĂŶƚ\u0003ŇĂŶŬŝŶŐ\u0003ƐĞƋƵĞŶĐĞ͕\u0003ǁĞ\u0003ĮƌƐƚ\u0003\nĞǀĂůƵĂƚĞĚ\u0003ŚŽǁ\u0003ŝŶĐƌĞĂƐŝŶŐ\u0003ŶƵŵďĞƌƐ\u0003ŽĨ\u0003ŵŝƐŵĂƚĐŚĞƐ\u0003ĂīĞĐƚĞĚ\u0003ďŝŶĚŝŶŐ\u0003ƚŽ\u0003ƚŚĞ\u0003ĐŽŶƐĞŶƐƵƐ\u0003ƐĞƋƵĞŶĐĞ\u0003\nϱ഻-AACCGGTT-ϯ഻\u0003ǁŝƚŚ\u0003ƉƌŽďĞƐ\u0003ďĞĂƌŝŶŐ\u0003Ăůů\u0003ƉŽƐƐŝďůĞ\u0003ŵŝƐŵĂƚĐŚ\u0003ĐŽŵďŝŶĂƟŽŶƐ͕\u0003ŝŶ\u0003ƚƌŝplicate. Average \n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted April 18, 2026. ; https://doi.org/10.64898/2026.04.16.719077doi: bioRxiv preprint \n\n16 \n \n^E\u0004W\u0003ŝŶƚĞŶƐŝƟĞƐ\u0003ĚĞĐƌĞĂƐĞĚ\u0003ĂƐ\u0003ƚŚĞ\u0003ŶƵŵďĞƌ\u0003ŽĨ\u0003ŵŝƐŵĂƚĐŚĞƐ\u0003ŝŶĐƌĞĂƐĞĚ\u0003;&ŝŐƵƌĞ\u0003ϰ\u0012Ϳ͕\u0003ǁŝƚŚ\u0003Ă\u0003ƐŚĂƌƉ\u0003ůŽƐƐ\u0003\nŽĨ\u0003ĂĸŶŝƚLJ\u0003ďĞƚǁĞĞŶ\u0003zero and one mismatches and near- ĐŽŵƉůĞƚĞ\u0003ůŽƐƐ\u0003ƵƉŽŶ\u0003ŝŶƚƌŽĚƵĐƟŽŶ\u0003ŽĨ\u0003ĨŽƵƌ\u0003\nmismatches. This conforms to the results of the SSL between the central and one mismatch rings \n(Figure 3D). dŚĞƐĞ\u0003 ƌĞƐƵůƚƐ\u0003 ĐŽŶĮƌŵ\u0003 ƚŚĂƚ\u0003 'Z,>Ϯ\u0003 ďŝŶĚŝŶŐ\u0003 ŝƐ\u0003 ĚĞƉĞŶĚĞŶƚ\u0003 ŽŶ\u0003 ŵĂŝŶƚĞŶĂŶĐĞ\u0003 ŽĨ\u0003 ƚŚĞ\u0003\nĐŽŶƐĞŶƐƵƐ\u0003 ƐĞƋƵĞŶĐĞ\u0003 ĂŶĚ\u0003 ƚŚĂƚ\u0003 ƐĞƋƵĞŶƟĂů\u0003 ĚŝƐƌƵƉƟŽŶ\u0003 ŽĨ\u0003 ƚŚĞ\u0003 ŵŽƟĨ\u0003 ƌĞƐƵůƚƐ\u0003 ŝŶ\u0003 ĞŶĞƌŐĞƟĐ\u0003 ƉĞŶĂůƟĞƐ\u0003\nrather than complete loss of binding. \n \u0011ĂƐĞĚ\u0003ŽŶ\u0003ƚŚĞ\u0003ĚŝŵĞƌ\u0003ŵŽƟĨ\u0003;&ŝŐƵƌĞ\u0003ϰ\u0004Ϳ\u0003ĂŶĚ\u0003ďĞĐĂƵƐĞ\u0003ƚŚĞƌĞ\u0003ŝƐ\u0003ĞǀŝĚĞŶĐĞ\u0003ƚŚĂƚ\u0003ƚŚĞ\u0003'Z,>\u0003ĨĂŵŝůLJ\u0003\nŽĨ\u0003 ƚƌĂŶƐĐƌŝƉƟŽŶ\u0003 ĨĂĐƚŽƌƐ\u0003 ĐĂŶ\u0003 ďŝŶĚ\u0003 ĂƐ\u0003 ĚŝŵĞƌƐ (43) ͕ \u0003ǁ Ğ \u0003Ŷ Ğ dž ƚ \u0003Ğ dž Ă ŵ ŝ Ŷ Ğ Ě \u0003ƚ Ś Ğ \u0003Ğ ī Ğ Đ ƚ Ɛ \u0003Ž Ĩ \u0003ŵ ŝ Ɛ ŵ Ă ƚ Đ Ś Ğ Ɛ \u0003\nŝŶƚƌŽĚƵĐĞĚ\u0003 ŝŶƚŽ\u0003 ĚŝŵĞƌŝĐ\u0003 'Z,>Ϯ\u0003 ŵŽƟĨƐ\u0003 ƐĞƉĂƌĂƚĞĚ\u0003 ďLJ\u0003 Ă\u0003 ƐŝŶŐůĞ\u0003 ďĂƐĞ\u0003ƉĂŝƌ ͘\u0003 \u0012ŽŶƚƌŽů\u0003 ƉƌŽďĞƐ\u0003 ǁĞƌĞ\u0003\nĚ Ğ Ɛ ŝ Ő Ŷ Ğ Ě \u0003ƚ Ž \u0003Đ Ž Ŷ ƚ Ă ŝ Ŷ \u0003ŝ Ě Ğ Ŷ Ɵ Đ Ă ů \u0003Ž ƌ \u0003Ă Ɛ LJ ŵ ŵ Ğ ƚ ƌ ŝ Đ \u0003ŵ ŝ Ɛ ŵ Ă ƚ Đ Ś Ğ Ɛ \u0003ŝ Ŷ \u0003ƚ Ś Ğ \u0003ů ĞŌ\u0003 ĂŶĚͬŽƌ\u0003 ƌŝŐŚƚ\u0003 ďŝŶĚŝŶŐ\u0003 ƐŝƚĞƐ͕\u0003\nallowing comparison of ind ŝǀŝĚƵĂů\u0003 ĂŶĚ\u0003 ŵƵůƟƉůĞ\u0003 ƉĞƌŵƵƚĂƟŽŶƐ. As shown in Figure 4D, probes \nĐŽŶƚĂŝŶŝŶŐ\u0003ŵĂƚĐŚĞĚ\u0003ŵŝƐŵĂƚĐŚĞƐ\u0003ŝŶ\u0003ďŽƚŚ\u0003ƐŝƚĞƐ\u0003ĚŝƐƉůĂLJĞĚ\u0003ƐƵďƐƚĂŶƟĂůůLJ\u0003ƌĞĚƵĐĞĚ\u0003ďŝŶĚŝŶŐ\u0003ƌĞůĂƟǀĞ\u0003ƚŽ\u0003\nƚŚĞ\u0003ĨƵůůLJ\u0003ĐŽŶƐĞŶƐƵƐ\u0003ĚŝŵĞƌ ͘\u0003,ŽǁĞǀĞƌ ͕\u0003ƉƌŽďĞƐ\u0003ǁŝƚŚ\u0003ŵŝƐŵĂƚĐŚĞƐ\u0003ĐŽŶĮŶĞĚ\u0003ƚŽ\u0003Ă\u0003ƐŝŶŐůĞ\u0003ƐŝƚĞ\u0003ƌĞƚĂŝŶĞĚ\u0003\niŶƚĞƌŵĞĚŝĂƚĞ\u0003ĂĸŶŝƚLJ͘\u0003These data suggest Ă\u0003ĐŽŽƉĞƌĂƟǀĞ\u0003ĐŽŶƚƌŝďƵƟŽŶ\u0003ŽĨ\u0003ƚŚĞ\u0003ƉĂŝƌĞĚ\u0003ƐŝƚĞƐ\u0003ƚŽ\u0003ŽǀĞƌĂůů\u0003\nGRHL2 binding. \n The ĐŽŶƚƌŝďƵƟŽŶ\u0003ŽĨ\u0003ĞĂĐŚ\u0003ƉŽƐƐŝďůĞ\u0003ƐŝŶŐůĞ-ďĂƐĞ\u0003ƐƵďƐƟƚƵƟŽŶ\u0003ǁŝƚŚŝŶ\u0003ƚŚĞ\u0003ĐŽŶƐĞŶƐƵƐ\u0003ƐŝƚĞ was \nthen analyzed. Consistent with the core ŵŽƟĨ\u0003ƐƉĞĐŝĮĐŝƚLJ\u0003ĚĞƐĐƌŝďĞĚ\u0003ĂďŽǀĞ͕\u0003ƐŝŶŐůĞ\u0003ŵŝƐŵĂƚĐŚĞƐ\u0003at \nƉŽƐŝƟŽŶƐ\u0003ϯ\u0003ĂŶĚ\u0003ϲ\u0003of the central \u0012dždž'\u0003resulted in ƚŚĞ\u0003ƐƚƌŽŶŐĞƐƚ\u0003ƌĞĚƵĐƟŽŶ\u0003ŝŶ\u0003ďŝŶĚŝŶŐ͘\u0003^ƵďƐƟƚƵƟŽŶƐ\u0003\nat the ƚĞƌŵŝŶĂů\u0003\u0004ͬd\u0003ƉŽƐŝƟŽŶƐ\u0003ǁĞƌĞ\u0003ŵŽƌĞ\u0003ƚŽůĞƌĂƚĞĚ\u0003;&ŝŐƵƌĞ\u0003ϰ\u001c and 4F). When all mismatch classes \n;ƵƉ\u0003 ƚŽ\u0003 ĨŽƵƌ\u0003 ƐƵďƐƟƚƵƟŽŶƐͿ\u0003 ǁĞƌĞ\u0003 ƌĂŶŬĞĚ\u0003 ďLJ\u0003 ^E\u0004W\u0003 ŝŶƚĞŶƐŝƚLJ͕\u0003 ƚŚĞ\u0003 ǀĂƌŝĂŶƚƐ with ŚŝŐŚĞƌ\u0003 ĂĸŶŝƚLJ \nmaintained the central \u0012dždž'\u0003 ĞůĞŵĞŶƚ\u0003 ĂŶĚ\u0003 ĚŝīĞƌĞĚ\u0003 ŵĂŝŶůLJ\u0003 Ăƚ\u0003 ĚŝƐƚĂů\u0003 ƉŽƐŝƟŽŶƐ͕\u0003 ǁŚŝůĞ\u0003 ƚŚĞ\u0003lower \nĂĸŶŝƚLJ\u0003ǀĂƌŝĂŶƚƐ\u0003ĚŝƐƌƵƉƚĞĚ\u0003ŽŶĞ\u0003Žƌ\u0003ŵŽƌĞ\u0003ďĂƐĞƐ\u0003ǁŝƚŚŝŶ\u0003ƚŚĞ\u0003ĐŽƌĞ\u0003;&ŝŐƵƌĞ\u0003ϰ&Ϳ͘ /ŶƚĞƌĞƐƟŶŐůLJ͕\u0003ŵƵƚĂƟŽŶƐ\u0003\n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted April 18, 2026. ; https://doi.org/10.64898/2026.04.16.719077doi: bioRxiv preprint \n\n17 \n \nŝŶ\u0003 ƚŚĞ\u0003 ƉŽƐŝƟŽŶƐ\u0003 ŇĂŶŬŝŶŐ\u0003 ƚŚĞ\u0003central CCGG resulted in ŚŝŐŚĞƌ\u0003 ŝŶƚĞŶƐŝƚLJ\u0003 ƚŚĂŶ\u0003 ŵƵƚĂƟŽŶƐ\u0003 ŝŶ\u0003 ƚŚĞ\u0003\nmiddle CG (ƉŽƐŝƟŽŶƐ\u00034-5) as was observed in Figure 3A-B. This may be due to the fact that Figure \n3 was based on genomic sequences, while the control sequences of Figure 4 evaluated all GRHL2 \nƉĞƌŵƵƚĂƟŽŶƐ with ƚŚĞ\u0003 ŝĚĞŶƟĐĂů\u0003 ŇĂŶŬŝŶŐ\u0003 ƐĞƋƵĞŶĐĞƐ. dŚĞƐĞ\u0003 ŽďƐĞƌǀĂƟŽŶƐ\u0003 ƌĞŝŶĨŽƌĐĞ\u0003 ƚŚĂƚ\u0003 'Z,>Ϯ\u0003\nƐƉĞĐŝĮĐŝƚLJ\u0003ŝƐ\u0003ĚŽŵŝŶĂƚĞĚ\u0003ďLJ\u0003ƌĞĐŽŐŶŝƟŽŶ\u0003ŽĨ\u0003ƚŚĞ\u0003central CdždžG͕\u0003ǁŝƚŚ\u0003ŇĂŶŬŝŶŐ\u0003ĂŶĚ\u0003ĞdžƚĞŶĚĞĚ\u0003ƉŽƐŝƟŽŶƐ\u0003\nĂĐƟŶŐ\u0003ŵŽƌĞ\u0003ĂƐ\u0003ŵŽĚƵůĂƚŽƌƐ\u0003ŽĨ\u0003ĂĸŶŝƚy. \n \nDoƟĨ sƉacing reǀeals Ɖeriodic diŵer-deƉendent binding and orientaƟon ƉreĨerences \n dŽ\u0003 ĞdžĂŵŝŶĞ\u0003the biophysical nature of GRHL2 dimers, we further evaluated the control \nƉƌŽďĞƐ\u0003 ĐŽŶƚĂŝŶŝŶŐ\u0003 ƚǁŽ\u0003 'Z,>Ϯ\u0003 ŵŽƟĨƐ\u0003 ƐĞƉĂƌĂƚĞĚ\u0003 ďLJ\u0003 ŝŶĐƌĞĂƐŝŶŐ\u0003 ŶƵŵďĞƌ Ɛ \u0003Ž Ĩ \u0003ď Ă Ɛ Ğ \u0003Ɖ Ă ŝ ƌ Ɛ ͘ \u0003t Ğ \u0003\nĞ dž Ă ŵ ŝ Ŷ Ğ Ě \u0003ƚ ǁ Ž \u0003Đ Ž Ŷ Į Ő Ƶ ƌ Ă Ɵ Ž Ŷ Ɛ ͗ \u0003Ă \u0003Ĩ Ž ƌǁ Ă ƌ Ě-ĨŽƌǁĂƌĚ\u0003 ŵŽƟĨ\u0003 ;\u0004\u0004\u0012\u0012''dd-AACCGGTT) and a reverse-\nforward arrangement (TTGGCCAA-\u0004\u0004\u0012\u0012''ddͿ͘\u0003 &ůƵŽƌĞƐĐĞŶĐĞ\u0003 ŝŶƚĞŶƐŝƚLJ\u0003 ǀĂƌŝĞĚ\u0003 ĂƐ\u0003 Ă\u0003 ĨƵŶĐƟŽŶ\u0003 ŽĨ\u0003\nƐƉĂĐŝŶŐ͕\u0003ĂŶĚ\u0003ƚŚĞ\u0003ƉĂƩĞƌŶ\u0003ǁĂƐ\u0003ĚŝƐƟŶĐƚ\u0003ďĞƚǁĞĞŶ\u0003ƚŚĞ\u0003ƚǁŽ\u0003ŽƌŝĞŶƚĂƟŽŶƐ\u0003;&ŝŐƵƌĞ\u0003ϱ\u0004Ϳ͘\u0003/Ŷ\u0003ƚŚĞ\u0003ĨŽƌǁĂƌĚ–\nĨŽƌǁĂƌĚ\u0003ĐŽŶĮŐƵƌĂƟŽŶ͕\u0003ďŝŶĚŝŶŐ\u0003ǁĂƐ\u0003ŚŝŐŚĞƐƚ\u0003ǁŚĞŶ\u0003ƚŚĞ\u0003ŵŽƟĨƐ\u0003ǁĞƌĞ\u0003ƉŽƐŝƟŽŶĞĚ\u0003ǁŝƚŚ\u0003Ă\u0003ϭ-base spacer, \nwith periodic decay as spacing increased. In contrast, the reverse-forward ŽƌŝĞŶƚĂƟŽŶ\u0003ƉƌĞĨĞƌƌĞĚ\u0003\na 3-4 bp spacer with less apparent periodic decay. \n Because the higher intensity was observed in the forward- ĨŽƌǁĂƌĚ\u0003ŵŽƟĨ͕\u0003ǁĞ\u0003ĂŶĂůLJnjĞĚ\u0003ƚŚŝƐ\u0003\nĐŽŶĮŐƵƌĂƟŽŶ\u0003 ŝŶ\u0003 ŵŽƌĞ\u0003 ĚĞƚĂŝů\u0003 ;&ŝŐƵƌĞ\u0003 ϱ\u0011Ϳ͘\u0003 WůŽƫŶŐ\u0003 ŇƵŽƌĞƐĐĞŶĐĞ\u0003 ŝŶƚĞŶƐŝƚLJ\u0003 ŽŶ\u0003 Ă\u0003 ůŽŐĂƌŝƚŚŵŝĐ\u0003 ƐĐĂůĞ\u0003\nĂŐĂŝŶƐƚ\u0003 ƐƉĂĐŝŶŐ\u0003 ƌĞǀĞĂůĞĚ\u0003 Ă\u0003 ƌĞƉĞĂƟŶŐ\u0003 ƉĂƩĞƌŶ͕\u0003 ǁŝƚŚ\u0003 ƉĞĂŬƐ\u0003 ŽĐĐƵƌƌŝŶŐ \u0003Ă ƚ \u0003ƌ Ğ Ő Ƶ ů Ă ƌ \u0003ŝ Ŷ ƚ Ğ ƌ ǀ Ă ů Ɛ ͘ \u0003\u0012 Ƶ ƌ ǀ Ğ \u0003\nĮƫŶŐ\u0003LJŝĞůĚĞĚ\u0003Ă\u0003ƉĞƌŝŽĚŝĐŝƚLJ\u0003ŽĨ ĂƉƉƌŽdžŝŵĂƚĞůLJ\u0003ϱ͘5 ďƉ͕\u0003ĐŽƌƌĞƐƉŽŶĚŝŶŐ\u0003ƚŽ\u0003ĂƉƉƌŽdžŝŵĂƚĞůLJ\u0003ŚĂůĨ\u0003ŽĨ\u0003ƚŚĞ\u0003\n10.5-bp helical repeat of B-form DNA (38). Data points were color-coded according to their phase \n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted April 18, 2026. ; https://doi.org/10.64898/2026.04.16.719077doi: bioRxiv preprint \n\n18 \n \nǁŝƚŚŝŶ\u0003ƚŚĞ\u0003ĮƩĞĚ\u0003ŚĞůŝĐĂů\u0003ƉŚĂƐŝŶŐ\u0003ŵŽĚĞů͕\u0003ĂŶĚ\u0003ƚŚŝƐ\u0003ƐŚŽǁĞĚ\u0003ƚŚĂƚ\u0003ŚŝŐŚĞƌ\u0003ďŝŶĚŝŶŐ\u0003ƚĞŶĚƐ\u0003ƚŽ\u0003ŽĐĐƵƌ\u0003ǁŚĞŶ\u0003\nƚŚĞ\u0003ƚǁŽ\u0003ŵŽƟĨƐ\u0003ĂůŝŐŶ\u0003ŽŶ\u0003ƚŚĞ\u0003ƐĂŵĞ\u0003ƌŽƚĂƟŽŶĂů\u0003ĨĂĐĞ\u0003ŽĨ\u0003ƚŚĞ\u0003ŚĞůŝdž͘ \n dŽ\u0003ǀŝƐƵĂůŝnjĞ\u0003ƚŚŝƐ\u0003ĞīĞĐƚ\u0003ŝŶ\u0003ƚĞƌŵƐ\u0003ŽĨ\u0003\u0018E\u0004\u0003ŐĞŽŵĞƚƌLJ͕\u0003ƚŚĞ\u0003ƐƉĂĐŝŶŐ\u0003ĂƌƌĂŶŐĞŵĞŶƚ\u0003ǁĂƐ\u0003ƉƌŽũĞĐƚĞĚ\u0003\nŽ Ŷ ƚ Ž \u0003Ă \u0003Ś Ğ ů ŝ Đ Ă ů \u0003ƌ Ğ Ɖ ƌ Ğ Ɛ Ğ Ŷ ƚ Ă Ɵ Ž Ŷ \u0003; & ŝ Ő Ƶ ƌ Ğ \u0003ϱ \u0012 Ϳ ͘ \u0003d Ś Ğ \u0003Đ Ž ů Ž ƌ Ğ Ě \u0003Ɖ Ğ Ă Ŭ \u0003Ɖ Ž Ɛ ŝƟŽŶƐ\u0003 ĐůƵƐƚĞƌĞĚ\u0003 ŽŶ\u0003 ĂůƚĞƌŶĂƟŶŐ\u0003\nĨĂĐĞƐ\u0003ŽĨ\u0003ƚŚĞ\u0003ŚĞůŝdž͕\u0003ŝŶĚŝĐĂƟŶŐ\u0003ƚŚĂƚ\u0003'Z,>Ϯ\u0003ĚŝŵĞƌƐ\u0003ďŝŶĚ\u0003ŵŽƐƚ\u0003ƐƚƌŽŶŐůLJ\u0003ǁŚĞn both sites fall on the \nƐĂŵĞ\u0003ŚĞůŝĐĂů\u0003ƐƵƌĨĂĐĞ\u0003ƌĂƚŚĞƌ\u0003ƚŚĂŶ\u0003ďĞŝŶŐ\u0003ƌŽƚĂƚĞĚ\u0003ĂǁĂLJ\u0003ĨƌŽŵ\u0003ŽŶĞ\u0003ĂŶŽƚŚĞƌ ͘\u0003dŚŝƐ\u0003ƉĂƩĞƌŶ\u0003ƐƵƉƉŽƌƚƐ\u0003Ă\u0003\nŵŽĚĞů\u0003ŝŶ\u0003ǁŚŝĐŚ\u0003ƐƉĂƟĂů\u0003ƉŽƐŝƟŽŶŝŶŐ\u0003ĂůŽŶŐ\u0003ƚŚĞ\u0003ŚĞůŝdž\u0003ĨĂĐŝůŝƚĂƚĞƐ\u0003ƐŝŵƵůƚĂŶĞŽƵƐ\u0003Žƌ\u0003ĐŽŽƉĞƌĂƟǀĞ\u0003ďŝŶĚŝŶŐ\u0003\nof two GRHL2 molecules. \n \u0004\u0003ƐƚƌƵĐƚƵƌĂů\u0003ŵŽĚĞů\u0003ǁĂƐ\u0003ŐĞŶĞƌĂƚĞĚ\u0003ƚŽ\u0003ŝůůƵƐƚƌĂƚĞ\u0003ƚŚĞ\u0003'Z,>Ϯ\u0003ĐŽŶĮŐƵƌĂƟŽŶ\u0003ƉƌĞĚŝĐƚĞĚ\u0003Ăƚ\u0003ƚŚĞ\u0003\nŽƉƟŵĂů\u0003 ƐƉĂĐŝŶŐ\u0003 ĚŝƐƚĂŶĐĞ\u0003 ;&ŝŐƵƌĞ\u0003 ϱ\u0018͕\u0003 ^ƵƉƉůĞŵĞŶƚĂry Movie 1). Using the published crystal \nstructure of a GRHL1 single dimer (comprised of two monomers) on a single DNA ŵŽƟĨ as a \ntemplate (43), two GRHL2 dimers (each comprised of two monomers) ǁĞƌĞ\u0003ƉŽƐŝƟŽŶĞĚ\u0003ŽŶ\u0003ƉĂŝƌĞĚ\u0003\nŵŽƟĨƐ\u0003ƐĞƉĂƌĂƚĞĚ\u0003ďLJ\u0003each spacing of increasing base pair s using PyMol. The intervening B-DNA \nĮďĞƌ\u0003ƐƚƌƵĐƚƵƌĞƐ\u0003ƵƐĞĚ\u0003ƚŽ\u0003ĚĞĮŶĞ\u0003ƉƌĞĐŝƐĞ\u0003ŚĞůŝĐĂů\u0003ŐĞŽŵĞƚƌLJ\u0003ĂŶĚ\u0003ƐƉĂĐŝŶŐ\u0003ǁĞƌĞ\u0003ŐĞŶĞƌĂƚĞĚ\u0003ǁŝƚŚ\u0003tĞď\u0003\n3DNA 2.0 (47). At 1-bp spacing, the two sets of GRHL2 dimers are oriented on the same side of \nƚŚĞ\u0003 \u0018E\u0004\u0003 ŝŶ\u0003 ĐůŽƐĞ\u0003 ƉƌŽdžŝŵŝƚLJ (Figure 5D). These results suggest that GRHL2 can bind dimeric \n\u0004\u0004\u0012\u0012''dd\u0003 ƐŝƚĞƐ\u0003 ǁŝƚŚ\u0003 ƉƌĞĨĞƌĞŶĐĞ\u0003 ĨŽƌ\u0003 ƚŚŽƐĞ\u0003 ƚŚĂƚ\u0003 ĂƌĞ\u0003 ĂƌƌĂŶŐĞĚ\u0003 ŝŶ\u0003 ƚŚĞ\u0003 ƐĂŵĞ\u0003 ŽƌŝĞŶƚĂƟŽŶ\u0003 ĂŶĚ\u0003\nƉŽƐŝƟŽŶĞĚ\u0003 ŽŶ\u0003 ƚŚĞ\u0003 ƐĂŵĞ\u0003 ŚĞůŝĐĂů\u0003 ĨĂĐĞ͕\u0003 ǁŚŝĐŚ\u0003 ŵĂLJ\u0003 ƌĞŇĞĐƚ\u0003 ŐĞŽŵĞƚƌŝĐ\u0003ĐŽŶƐƚƌĂŝŶƚƐ\u0003 ŽĨ\u0003 ĚŝŵĞƌ-dimer \nŝŶƚĞƌĂĐƟŽŶƐ\u0003ŽŶ\u0003\u0018E\u0004͘ \n \nDisƟnguisŚing direct GRHL2 binding Ĩroŵ indirect, coĨactor-mediated recruitment across genomic \nloci \n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted April 18, 2026. ; https://doi.org/10.64898/2026.04.16.719077doi: bioRxiv preprint \n\n19 \n \n The GRHL2 genomic SNAP array ǁĂƐ\u0003 ŶĞdžƚ\u0003 ƵƐĞĚ\u0003to discriminate genomic regions where \nGRHL2 binds DNA directly from those where ChIP-ƐĞƋ\u0003ĞŶƌŝĐŚŵĞŶƚ\u0003ůŝŬĞůLJ\u0003ƌĞŇĞĐƚƐ\u0003indirect, protein-\nƚĞƚŚĞƌĞĚ\u0003ŝŶƚĞƌĂĐƟŽŶƐ\u0003;Figure 6A). If a genomic region showed GRHL2 occupancy by ChIP-seq and \nalso produced a detectable signal on the SNAP array, the site ǁĂƐ\u0003 ĐůĂƐƐŝĮĞĚ\u0003as a direct DNA-\nbinding event. In contrast, regions with strong ChIP-seq signal but no measurable binding on the \nSNAP array were considered indirect binding events, consistent with GRHL2 being tethered \nƚŚƌŽƵŐŚ\u0003ĂŶ\u0003ŝŶƚĞƌĂĐƟŶŐ\u0003ƚƌĂŶƐĐƌŝƉƟŽŶ\u0003ĨĂĐƚŽƌ\u0003or cofactor rather than ĐŽŶƚĂĐƟŶŐ DNA directly at the \nŵŽƟĨ. ZĞƉƌĞƐĞŶƚĂƟǀĞ\u0003ĞdžĂŵƉůĞƐ\u0003ŽĨ\u0003ĞĂĐŚ\u0003ĐĂƚĞŐŽƌLJ\u0003of binding event are shown in Figure 6B. At the \nMCTS1 promoter, GRHL2 ChIP-seq peaks were  observed under both endogenous and \nŽǀĞƌĞdžƉƌĞƐƐĞĚ\u0003'Z,>Ϯ\u0003ĐŽŶĚŝƟŽŶƐ (7, 30), and the SNAP array revealed a clear binding signal across \na subset of adjacent Ɵ ů Ğ Ě \u0003Ɖ ƌ Ž ď Ğ Ɛ \u0003ŝ Ŷ \u0003ƚ Ś ŝ Ɛ \u0003ƌ Ğ Ő ŝ Ž Ŷ that represent a single GRHL2 binding site \nĂƉƉƌŽdžŝŵĂƚĞůLJ\u0003at the center of the ChIP-seq peak͕\u0003ŝŶĚŝĐĂƟŶŐ\u0003ĚŝƌĞĐƚ\u0003ƌĞĐŽŐŶŝƟŽŶ\u0003ŽĨ\u0003ƵŶĚĞƌůLJŝŶŐ\u0003\u0018E\u0004\u0003\nsequences. At the EPIC1 promoter, GRHL2 was also enriched by ChIP-seq. However, SNAP probes \nƐƉĂŶŶŝŶŐ\u0003ƚŚĞ\u0003ƐĂŵĞ\u0003ŝŶƚĞƌǀĂů\u0003ƐŚŽǁĞĚ\u0003ůŝƩůĞ\u0003ƚŽ\u0003ŶŽ\u0003ďŝŶĚŝŶŐ\u0003ƐŝŐŶĂů͕\u0003ĐŽŶƐŝƐƚĞŶƚ\u0003ǁŝƚŚ\u0003'Z,>Ϯ\u0003ůŽĐĂůŝnjĂƟŽŶ\u0003\nbeing mediated by another DNA-bound factor at this site. \n Across the full genomic SNAP dataset, 772,732 probes collapsed into 6,151 ĚŝƐƟŶĐƚ\u0003\nŐĞŶŽŵŝĐ\u0003ƌĞŐŝŽŶƐ\u0003ďĂƐĞĚ\u0003ŽŶ\u0003ƟůŝŶŐ\u0003ŽǀĞƌůĂƉ\u0003;&ŝŐƵƌĞ\u0003ϲ\u0012Ϳ͘\u0003hƐŝŶŐ\u0003ƚŚĞ\u0003ŚŝŐŚĞƐƚ-intensity probe within each \nƌĞŐŝŽŶ\u0003ĂƐ\u0003ƚŚĞ\u0003ƌĞƉƌĞƐĞŶƚĂƟǀĞ\u0003ǀĂůƵĞ͕\u0003ϰ͕ϱϲϲ\u0003ƌĞŐŝŽŶƐ\u0003ǁĞƌĞ\u0003ĐůĂƐƐŝĮĞĚ\u0003ĂƐ\u0003ĚŝƌĞĐƚ\u0003'Z,>Ϯ\u0003ďŝŶĚŝŶŐ\u0003ƐŝƚĞƐ \n(which includes perfect and mismatch sites), while 1,585 regions showed ChIP-seq signal without \nSNAP binding and were therefore assigned as indirect sites. These data indicate that Ϯϱ͘ϴй of \nGRHL2 ChIP-seq peaks ƟůĞĚ\u0003ŽŶ\u0003ƚŚĞ\u0003^E\u0004W\u0003ĂƌƌĂy correspond to indirect recruitment events rather \nƚŚĂŶ\u0003ĚŝƌĞĐƚ\u0003ƌĞĐŽŐŶŝƟŽŶ ŽĨ\u0003ƚŚĞ\u0003\u0018E\u0004\u0003ŵŽƟĨ. As ERɲ\u0003ĂŶĚ\u0003&Ždž\u0004ϭ\u0003ĂƌĞ\u0003ƚƌĂŶƐĐƌŝƉƟŽŶ\u0003ĨĂĐƚŽƌƐ\u0003reported to \n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted April 18, 2026. ; https://doi.org/10.64898/2026.04.16.719077doi: bioRxiv preprint \n\n20 \n \nŚĂ ǀ Ğ\u0003Ěŝƌ ĞĐƚ\u0003ŝŶƚĞƌ ĂĐƟŽŶƐ\u0003ǁŝƚŚ\u0003'Z,>Ϯ͕\u0003ƚŚĞ\u0003ϭ͕ϱϴϱ\u0003ŝŶĚŝƌ ĞĐƚ\u0003ďŝŶĚŝŶŐ\u0003ƌĞŐŝŽŶƐ\u0003 ǁĞƌĞ\u0003 ĐŽŵƉĂƌĞĚ\u0003 ƚŽ\u0003 ƚŚĞ\u0003\n4,566 direct binding regions in terms of both the genomic sequence and ChIP-seq datasets for \nƚŚĞƐĞ\u0003ƚƌĂŶƐĐƌŝƉƟŽŶ\u0003ĨĂĐƚŽƌƐ (29, 48, 49). ERɲ\u0003did not ŚĂǀĞ\u0003ĂŶLJ\u0003ƐƚĂƟƐƟĐĂů\u0003ĚŝīĞƌĞŶĐĞ in ƚŚĞ\u0003ƌĞůĂƟǀĞ\u0003\nnumber of regions that contain a genomic ERɲ\u0003half-site (5’-RGGTCA-3' half-site where R = A or G) \nnor in the ƌĞůĂƟǀĞ\u0003ŶƵŵďĞƌ\u0003ŽĨ\u0003ƌĞŐŝŽŶƐ\u0003ƚŚĂƚ\u0003ĐŽŶƚĂŝŶ\u0003a ChIP-seq peak from ƚǁŽ\u0003ĚŝīĞƌĞŶƚ\u0003studies of \nERɲ\u0003in the MCF7 breast cancer cell line (Supplementary Table 1). The ER ɲ\u0003ChIP-seq data were \nconducted with ĂŶƟďŽĚŝĞƐ\u0003ĂŐĂŝŶƐƚ\u0003total ERɲ\u0003protein (49) or ƐƉĞĐŝĮcally against phosphorylated \nERɲ\u0003at S118 (29). /Ŷ\u0003ĐŽŶƚƌĂƐƚ͕\u0003ƚŚĞ\u0003ŝŶĚŝƌĞĐƚ\u0003ƌĞŐŝŽŶƐ\u0003ĞdžŚŝďŝƚĞĚ\u0003Ă\u0003ŚŝŐŚĞƌ\u0003ƌĞůĂƟǀĞ\u0003ƉƌŽƉŽƌƟŽŶ of &Ždž\u0004ϭ\u0003\nChIP-seq binding sites as compared to direct regions͕\u0003ƐƵŐŐĞƐƟŶŐ\u0003ƚŚĂƚ\u0003recruitment of GRHL2 to the \nindirect regions may be driven, in part, by protein tethering ǁŝƚŚ\u0003&Ždž\u0004ϭ\u0003ƌĂƚŚĞƌ\u0003ƚŚĂŶ\u0003\u001cZɲ. \n The number of ĚŝƐƟŶĐƚ\u0003direct binding events occurring ǁŝƚŚŝŶ\u0003ĞĂĐŚ\u0003ŐĞŶŽŵŝĐ\u0003ŝŶƚĞƌǀĂů\u0003ƟůĞĚ\u0003\non the array was then determined. Most regions contained either one or two ĚŝƐƟŶĐƚ\u0003 ĚŝƌĞĐƚ\u0003\nGRHL2 binding sites, with a smaller number containing three or more (Figure 6D). Overall, these \nresults demonstrate that combining SNAP array measurements with genome-wide ChIP-seq \nstudies ĞŶĂďůĞƐ\u0003ĨƵŶĐƟŽŶĂů\u0003ƐĞƉĂƌĂƟŽŶ\u0003ŽĨ\u0003ĚŝƌĞĐƚ\u0003ƐĞƋƵĞŶĐĞ-encoded GRHL2 binding from cofactor-\nĚĞƉĞŶĚĞŶƚ\u0003ŽĐĐƵƉĂŶĐLJ͕\u0003ƌĞǀĞĂůŝŶŐ\u0003ĚŝƐƟŶĐƚ\u0003ĐůĂƐƐĞƐ\u0003ŽĨ\u0003'Z,>Ϯ-associated genomic loci. \n \nA dense cluster oĨ GRHL2 moƟĨs corresƉonds to a ŚigŚly occuƉied CŚIP-seq region \n /Ŷ\u0003 ĞdžĂŵŝŶŝŶŐ\u0003 ŚŽǁ\u0003 ŽŌĞŶ\u0003 ƚŚĞ\u0003 'Z,>Ϯ\u0003 ďŝŶĚŝŶŐ\u0003 ŵŽƟĨ\u0003 ŽĐĐƵƌƐ\u0003 ĂĐƌŽƐƐ\u0003 ƚŚĞ\u0003ŐĞŶŽŵĞ͕\u0003 Ă\u0003low-\nĐŽŵƉůĞdžŝƚLJ\u0003region on chromosome 7 that contains an unusually high density of GRHL2 consensus \nŵŽƟĨƐ\u0003 ǁĂƐ\u0003 ŝĚĞŶƟĮĞĚ\u0003 ;&ŝŐƵƌĞ\u0003 ϳ\u0004Ϳ͘\u0003 tŝƚŚŝŶ\u0003 ƚŚŝƐ\u0003 ŝŶƚĞƌǀĂů\u0003 ĨƌŽŵ\u0003 ĐŚƌϳ͗ϭϱϱ͕ϭϮϰ͕ϬϬϰ-155,124,759, \nŵƵůƟƉůĞ\u0003 ĞdžĂĐƚ\u0003 ĐĂŶŽŶŝĐĂů\u0003 \u0004\u0004\u0012\u0012''dd\u0003 ŵŽƟĨƐ\u0003 ǁĞƌĞ\u0003 ƉƌĞƐĞŶƚ͕\u0003 ŝŶƚĞƌƐƉĞƌƐĞĚ\u0003 ǁŝƚŚ\u0003 ĂĚĚŝƟŽŶĂů\u0003 ƐŝŶŐůĞ-\n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted April 18, 2026. ; https://doi.org/10.64898/2026.04.16.719077doi: bioRxiv preprint \n\n21 \n \nmismatch variants. This arrangement created a concentrated cluster of closely spaced GRHL2 \nƌĞĐŽŐŶŝƟŽŶ\u0003ƐŝƚĞƐ\u0003ĂĐƌŽƐƐ\u0003Ă\u0003ƌĞůĂƟǀĞůLJ\u0003ƐŚŽƌƚ\u0003ƐƚƌĞƚĐŚ\u0003ŽĨ\u0003\u0018E\u0004͘\u0003\u0012Ś/W-seq data from our group (7, 30) \nshowed strong GRHL2 enrichment at the edges of this region under both datasets (Figure 7B, \nƚŽƉͿ͘\u0003tŚĞŶ\u0003ŵĂƉƉĞĚ\u0003ŽŶ\u0003ƚŚĞ\u0003^E\u0004W\u0003ĂƌƌĂLJ͕\u0003ƉƌŽďĞƐ\u0003ƟůĞĚ\u0003ĞǀĞƌLJ\u0003Ɛŝdž\u0003ďĂƐĞ\u0003ƉĂŝƌƐ\u0003ĂĐƌŽƐƐ\u0003ƚŚĞ\u0003ƐĂŵĞ\u0003ŝŶƚĞƌǀĂů\u0003\nproduced a broad band of elevated binding signal within the central region ;&ŝŐƵƌĞ\u0003ϳ\u0011͕\u0003ďŽƩŽŵͿ͘\u0003\nZĂƚŚĞƌ\u0003ƚŚĂŶ\u0003Ă\u0003ƐŝŶŐůĞ\u0003ĚŝƐĐƌĞƚĞ\u0003ďŝŶĚŝŶŐ\u0003ƉĞĂŬ͕\u0003ƚŚĞ\u0003^E\u0004W\u0003ƐŝŐŶĂů\u0003ŇƵĐƚƵĂƚĞĚ\u0003ĂĐƌŽƐƐ\u0003ƚŚĞ\u0003ƟůĞĚ\u0003ƉƌŽďĞƐ͕\u0003\nǁŚŝĐŚ\u0003ŝƐ\u0003ĐŽŶƐŝƐƚĞŶƚ\u0003ǁŝƚŚ\u0003ŵƵůƟƉůĞ\u0003ďŝŶĚŝŶŐ\u0003ĞǀĞŶƚƐ\u0003ĂƌŝƐŝŶŐ\u0003ĨƌŽŵ\u0003ƚŚĞ\u0003ƌĞƉĞĂƚĞĚ\u0003ĐŽŶƐĞŶƐƵƐ\u0003ĂŶĚ\u0003ŶĞĂƌ-\nĐŽŶƐĞŶƐƵƐ\u0003 ŵŽƟĨƐ͘\u0003 dŚĞ\u0003 ŽďƐĞƌǀĞĚ\u0003 ĚŝīĞƌĞnce between the ChIP-seq and SNAP data could be \nĞdžƉůĂŝŶĞĚ\u0003ďLJ\u0003ŝŵƉƌĞĐŝƐĞ\u0003ŵĂƉƉŝŶŐ\u0003ŽĨ\u0003\u0012Ś/W-seq reads in this low-ĐŽŵƉůĞdžŝƚLJ\u0003ƌĞŐŝŽŶ\u0003ƚŚĂƚ\u0003ƌĞƐƵůƚed in \nƚǁŽ\u0003ƉĞĂŬƐ\u0003ŽŶ\u0003ĞŝƚŚĞƌ\u0003ƐŝĚĞ\u0003ŽĨ\u0003ƚŚĞ\u0003ƌĞŐŝŽŶ͘\u0003/Ŷ\u0003ƉĂƌƟĐƵůĂƌ͕\u0003ƚŚĞ\u0003\u0012Ś/W-seq peak on one side of the region \nmay represent the perfect GRHL2 consensus (5’-AACCGGTT-3’) reads while the ChIP-seq peak on \nother side of the region may represent the primary mismatch site (5’-AAACGGTT-3’) reads. \n\u0004ůƚĞƌŶĂƟǀĞůLJ͕\u0003ƚŚŝƐ\u0003ƌĞŐŝŽŶ\u0003ŵĂLJ\u0003ďĞ\u0003ĐŚƌŽŵĂƟŶ\u0003ŝŶĂĐĐĞƐƐŝďůĞ\u0003ƚŽ\u0003ƚŚĞ\u0003\u0012Ś/W-ƐĞƋ\u0003ĂŶƟďŽĚLJ\u0003ƌĞĂŐĞŶƚƐ͘\u0003These \nŽďƐĞƌǀĂƟŽŶƐ\u0003ŝŶĚŝĐĂƚĞ\u0003ƚŚĂƚ\u0003ƚŚŝƐ\u0003ƌĞŐŝŽŶ\u0003ƉŽƚĞŶƟĂůůLJ\u0003represents a highly occupied GRHL2 locus driven \nďLJ\u0003ƚŚĞ\u0003ƉƌĞƐĞŶĐĞ\u0003ŽĨ\u0003ŵĂŶLJ\u0003ĐůŽƐĞůLJ\u0003ƐƉĂĐĞĚ\u0003ďŝŶĚŝŶŐ\u0003ƐŝƚĞƐ͘\u0003dŚĞ\u0003ĐŽŵďŝŶĂƟŽŶ\u0003ŽĨ\u0003ĚĞŶƐĞ\u0003ĐŽŶƐĞŶƐƵƐ\u0003ŵŽƟĨƐ\u0003\nand favorable single-mismatch variants likely creates a local DNA environment that strongly favors \n'Z,>Ϯ\u0003ďŝŶĚŝŶŐ\u0003ĂĐƌŽƐƐ\u0003ƚŚĞ\u0003ĞŶƟƌĞ\u0003ŝŶƚĞƌǀĂů͘ \n\u0003\n\u0018/^\u0012h^^/KE\u0003\n In this study, we used a customized high-density genomic SNAP DNA-binding array to \nbiochemically ĚĞĮŶĞ͕\u0003 Ăƚ\u0003 Ɛŝdž-ďĂƐĞ\u0003 ƌĞƐŽůƵƟŽŶ͕\u0003 ƚŚĞ\u0003 ƐĞƋƵĞŶĐĞ\u0003 ĨĞĂƚƵƌĞƐ\u0003 ƚŚĂƚ\u0003 ŐŽǀĞƌŶ\u0003 'Z,>Ϯ\u0003 \u0018E\u0004\u0003\nƌĞĐŽŐŶŝƟŽŶ\u0003 ĂĐƌŽƐƐ\u0003 ƚŚĞ\u0003 ŐĞŶŽŵĞ͘\u0003 \u0011LJ\u0003 ŝŶƚĞŐƌĂƟŶŐ\u0003 ĚŝƌĞĐƚ\u0003 ďŝŶĚŝŶŐ\u0003 ŵĞĂƐƵƌĞŵĞŶƚƐ\u0003 ǁŝƚŚ\u0003 ƐƚƌƵĐƚƵƌĂů\u0003\n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted April 18, 2026. ; https://doi.org/10.64898/2026.04.16.719077doi: bioRxiv preprint \n\n22 \n \nŝŶƚĞƌƉƌĞƚĂƟŽŶ\u0003ĂŶĚ\u0003\u0012Ś/W-ƐĞƋ\u0003ĚĂƚĂ͕\u0003ǁĞ\u0003ĚŝƐƟŶŐƵŝƐŚ\u0003ŵŽƟĨ-encoded GRHL2 direct binding events from \ncofactor-dependent occupancy to ƌĞƐŽůǀĞ\u0003ŚŽǁ\u0003ŵŽƟĨ\u0003ĐŽŵƉŽƐŝƟŽŶ͕\u0003ŇĂŶŬŝŶŐ\u0003ƐĞƋƵĞŶĐĞ͕\u0003ĂŶĚ\u0003ŵŽƟĨ\u0003\nƐƉĂĐŝŶŐ\u0003ĐŽůůĞĐƟǀĞůLJ\u0003dictate GRHL2 binding ĐŚĂƌĂĐƚĞƌŝƐƟĐƐ. While prior studies have established \nGRHL2 as a co-ƌĞŐƵůĂƚŽƌ\u0003ŽĨ\u0003ƚƌĂŶƐĐƌŝƉƟŽŶ\u0003ƚŚƌŽƵŐŚ\u0003ŝŶƚĞƌĂĐƟŽŶƐ\u0003ǁŝƚŚ\u0003ĨĂĐƚŽƌƐ\u0003ƐƵĐŚ\u0003ĂƐ\u0003\u001cZɲ\u0003and PR and \nƚŚƌŽƵŐŚ\u0003ŵŽĚƵůĂƟŽŶ\u0003ŽĨ\u0003ĐŚƌŽŵĂƟŶ\u0003ĂĐĐĞƐƐŝďŝůŝƚLJ͕\u0003ƚŚĞƐĞ\u0003ĂƉƉƌŽĂĐŚĞƐ\u0003ĚŽ\u0003ŶŽƚ\u0003ĚŝīĞƌĞŶƟĂƚĞ direct DNA \nbinding from indirect, cofactor-mediated recruitm ent. The present work addresses this gap by \nĚĞĮŶŝŶŐ\u0003ƚŚĞ\u0003ŝŶƚƌŝŶƐŝĐ\u0003ƐĞƋƵĞŶĐĞ\u0003ĚĞƚĞƌŵŝŶĂŶƚƐ\u0003ŽĨ\u0003'Z,>Ϯ\u0003ďŝŶĚŝŶŐ\u0003ŝŶĚĞƉĞŶĚĞŶƚ\u0003ŽĨ\u0003ĐŚƌŽŵĂƟŶ\u0003ĐŽŶƚĞdžƚ \n(50, 51). \n Across more than 772,732 genomic probe sequences, GRHL2 binding prefers a conserved \n\u0004 \u0004 \u0012 \u0012 ' ' d d \u0003ŵ Ž Ɵ Ĩ ͘ \u0003^ Ƶ ď Ɛ Ɵ ƚ Ƶ Ɵ Ž Ŷ Ɛ \u0003Ă ƚ \u0003Ɖ Ž Ɛ ŝ Ɵons 3 (C) ĂŶĚ\u0003 ƉŽƐŝƟŽŶ\u0003 ϲ (G) produced the steepest \nƌĞĚƵĐƟŽŶƐ\u0003ŝŶ\u0003^E\u0004W\u0003ŝŶƚĞŶƐŝƚLJ. SƵďƐƟƚƵƟŽŶƐ\u0003Ăƚ\u0003ƚŚĞ\u0003central \u0012'\u0003;ƉŽƐŝƟŽŶƐ\u0003ϰ-5) had minor ĞīĞĐƚƐ, and \nƐƵďƐƟƚƵƟŽŶƐ\u0003Ăƚ\u0003ƚŚĞ\u0003ŽƵƚĞƌ\u0003\u0004ͬd\u0003ƉŽƐŝƟŽŶƐ\u0003ǁĞƌĞ\u0003ŵŽƌĞ\u0003ƚŽůĞƌĂƚĞĚ͘\u0003dŚĞƐĞ\u0003ƌĞƐƵůƚƐ\u0003ĂƌĞ consistent with \nƐƚƌƵĐƚƵƌĂů\u0003ƐƚƵĚŝĞƐ\u0003ŽĨ\u0003ƚŚĞ\u0003'ƌĂŝŶLJŚĞĂĚͬ\u0012WϮ\u0003ĨĂŵŝůLJ\u0003(43). TŚĞ\u0003ƌĞĐŽŐŶŝƟŽŶ\u0003ŚĞůŝdž\u0003ŽĨ\u0003ƚŚĞ\u0003ǁŝŶŐĞĚ-ŚĞůŝdž\u0003\nDNA-ďŝŶĚŝŶŐ\u0003 ĚŽŵĂŝŶ\u0003 ĐŽŶƚĂĐƚƐ\u0003 ƚŚĞ\u0003 ƉŽƐŝƟŽŶƐ\u0003 ϯ\u0003 ĂŶĚ\u0003 ϲ\u0003 ŽĨ\u0003 ƚŚĞ\u0003 \u0004\u0004\u0012\u0012''dd\u0003 ŵŽƟĨ\u0003 ǁŝƚŚ\u0003 ŶŽ\u0003 ĚŝƌĞĐƚ\u0003\ncontacts to ƚŚĞ\u0003ĐĞŶƚƌĂů\u0003\u0012'͕\u0003ƌĞƐƵůƟŶŐ\u0003ŝŶ a \u0012dždž'\u0003ĐŽƌĞ͘\u0003The limited backbone contacts outside of the \noctamer ĐŽƌĞ\u0003 ŵŽƟĨ\u0003found in the crystal structure (43) Ɛ Ƶ Ő Ő Ğ Ɛ ƚ \u0003ƚ Ś Ă ƚ \u0003ƚ Ś Ğ \u0003Ň Ă Ŷ Ŭ ŝ Ŷ Ő \u0003Ɖ ƌ Ğ Ĩ Ğ ƌ Ğ Ŷ Đ Ğ Ɛ \u0003\nuncovered in our analysis may be driven by DNA structural features, such as major groove \nŐĞŽŵĞƚƌLJ\u0003 ŝŶŇƵĞŶĐĞĚ\u0003 ďLJ\u0003 \u0004ͬd\u0003 ĐŽŶƚĞŶƚ͘ The SSL and mismatch analyses also ĞdžŚŝďŝƚĞĚ this \nenrichment of tolerated single-ďĂƐĞ\u0003ƐƵďƐƟƚƵƟŽŶƐ\u0003Ăƚ\u0003ƚŚĞ\u0003ŵŝĚĚůĞ\u0003\u0012'\u0003ƉŽƐŝƟŽŶƐ\u0003ŽĨ\u0003ƚŚĞ\u0003ŵŽƟĨ͕\u0003ǁŝƚŚ\u0003Ă\u0003\nƐƚƌŽŶŐ\u0003 ďŝĂƐ\u0003 ƚŽǁĂƌĚ\u0003 \u0012їdͬ\u0004\u0003 ĂŶĚ\u0003 'ї\u0004ͬd\u0003 ĐŚĂŶŐĞƐ͘\u0003 KŶĞ\u0003 ƉŽƐƐŝďůĞ\u0003 ĞdžƉůĂŶĂƟŽŶ\u0003 ŝƐ\u0003 ƚŚĂƚ\u0003 ƚŚĞƐĞ\u0003\nƐƵďƐƟƚƵƟŽŶƐ\u0003 ƉƌĞƐĞƌǀĞ\u0003 \u0018E\u0004\u0003 ƐŚĂƉĞ\u0003 Žƌ\u0003 ŚLJĚƌŽŐĞŶ-bonding geometry in a way that ŝƐ\u0003 ƉĂƌƟĂůůLJ\u0003\n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted April 18, 2026. ; https://doi.org/10.64898/2026.04.16.719077doi: bioRxiv preprint \n\n23 \n \nĐŽŵƉĂƟďůĞ\u0003ǁŝƚŚ\u0003'Z,>Ϯ\u0003ƌĞĐŽŐŶŝƟŽŶ͕\u0003ĂůůŽǁŝŶŐ\u0003ůŽǁ-ĂĸŶŝƚLJ\u0003ďŝŶĚŝŶŐ\u0003ƌĂƚŚĞƌ\u0003ƚŚĂŶ\u0003ƚŚĞ\u0003ĐŽŵƉůĞƚĞ\u0003ůŽƐƐ\u0003\nŽĨ\u0003ƚŚĞ\u0003ŝŶƚĞƌĂĐƟŽŶ͘\u0003 \n The SSL analysis provides ƋƵĂŶƟƚĂƟǀĞ\u0003detail towards a structural model in which the \ncentral \u0012dždž'\u0003ĂĐƚƐ\u0003ĂƐ\u0003ƚŚĞ\u0003ƉƌŝŶĐŝƉĂů\u0003ĞŶĞƌŐĞƟĐ\u0003contributor for GRHL2-\u0018E\u0004\u0003ƌĞĐŽŐŶŝƟŽŶ͕\u0003ǁŚŝůĞ\u0003ŇĂŶŬŝŶŐ\u0003\nƐĞƋƵĞŶĐĞƐ\u0003ĂŶĚ\u0003ƚŚĞ\u0003ĐĞŶƚƌĂů\u0003\u0012'\u0003ƚƵŶĞ\u0003ďŝŶĚŝŶŐ\u0003ĂĸŶŝƚLJ\u0003and ƐƉĞĐŝĮĐŝƚLJ. TŚŝƐ\u0003ůĞǀĞů\u0003ŽĨ\u0003ƌĞƐŽůƵƟŽŶ\u0003ĞdžƚĞŶĚƐ\u0003\nďĞLJŽŶĚ\u0003 ƚƌĂĚŝƟŽŶĂů\u0003 ŵŽƟĨ\u0003 ĞŶƌŝĐŚŵĞŶƚ\u0003 ĂŶĂůLJƐĞƐ o f  C h I P - s e q  p e a k s  ďLJ\u0003 ƉƌŽǀŝĚŝŶŐ\u0003 Ă\u0003 ƋƵĂŶƟƚĂƟǀĞ\u0003\nĚĞƐĐƌŝƉƟŽŶ\u0003ŽĨ\u0003ĞdžĂĐƚůLJ\u0003ǁŚĞƌĞ\u0003'Z,>Ϯ\u0003ŵĂLJ\u0003ďŝŶĚ\u0003ǁŝƚŚŝŶ\u0003a genomic region to a “sub-peak” level. This \nĞŶĂďůĞƐ\u0003 ĚŝƌĞĐƚ\u0003 ŝŶƚĞƌƉƌĞƚĂƟŽŶ\u0003 ŽĨ\u0003 ŐĞŶŽŵŝĐ\u0003 'Z,>Ϯ\u0003 ŽĐĐƵƉĂŶĐLJ\u0003to ĚŝƐƟŶŐƵŝƐh sequence-driven \nďŝŶĚŝŶŐ\u0003ĨƌŽŵ\u0003ŝŶĚŝƌĞĐƚ\u0003ƌĞĐƌƵŝƚŵĞŶƚ\u0003ŽďƐĞƌǀĞĚ\u0003ŝŶ\u0003ĐŚƌŽŵĂƟŶ-based assays.  \n dŚĞ\u0003ĞŶƌŝĐŚŵĞŶƚ\u0003ŽĨ\u0003ƚŽůĞƌĂƚĞĚ\u0003ƐƵďƐƟƚƵƟŽŶƐ\u0003Ăƚ\u0003ƚŚĞ\u0003ĐĞŶƚƌĂů\u0003\u0012'\u0003ƉŽƐŝƟŽŶƐ\u0003ĨƵƌƚŚĞƌ\u0003ƐƵŐŐĞƐƚƐ\u0003ƚŚĂƚ\u0003\n'Z,>Ϯ\u0003ďŝŶĚŝŶŐ\u0003ŵĂLJ\u0003ďĞ\u0003ƌĞůĂƟǀĞůLJ\u0003ƌŽďƵƐƚ\u0003ƚŽ\u0003ƐĞƋƵĞŶĐĞ\u0003ǀĂƌŝĂƟŽŶ\u0003Ăƚ\u0003\u0012Ɖ'\u0003ĚŝŶƵĐůĞŽƟĚĞƐ͕\u0003ǁŚŝĐŚ\u0003ĂƌĞ\u0003\nknown to be highly mutable in mamma lian genomes due to cytosine deamina ƟŽŶ (13). The \nŽďƐĞƌǀĞĚ\u0003ƚŽůĞƌĂŶĐĞ\u0003ƚŽ\u0003ƐƉĞĐŝĮĐ\u0003\u0012їdͬ\u0004\u0003ĂŶĚ\u0003'ї\u0004ͬd\u0003ƐƵďƐƟƚƵƟŽŶƐ\u0003brings forth the idea that GRHL2 \nďŝŶĚŝŶŐ\u0003ĐĂŶ\u0003ƉĞƌƐŝƐƚ͕\u0003Ăƚ\u0003ƌĞĚƵĐĞĚ\u0003ĂĸŶŝƚLJ͕\u0003ŝŶ\u0003ƌĞŐƵůĂƚŽƌLJ\u0003ƌĞŐŝŽŶƐ\u0003ƵŶĚĞƌŐŽŝŶŐ\u0003\u0012Ɖ'\u0003ĞƌŽƐŝŽŶ͘\u0003dŚŝƐ\u0003ŵĂLJ\u0003\nŚĂǀĞ\u0003ŝŵƉůŝĐĂƟŽŶƐ\u0003ĨŽƌ\u0003ƚŚĞ\u0003ĞǀŽůƵƟŽŶ\u0003ŽĨ\u0003'Z,>Ϯ\u0003ďŝŶĚŝŶŐ\u0003ƐŝƚĞƐ\u0003ĂŶĚ\u0003ĨŽƌ\u0003ƚŚĞ\u0003ŵĂŝŶƚĞŶĂŶĐĞ\u0003ŽĨ\u0003ƌĞŐƵůĂƚŽƌLJ\u0003\nĨƵŶĐƟŽŶ\u0003ŝŶ\u0003ŐĞŶŽŵŝĐ\u0003ƌĞŐŝŽŶƐ ƚŚĂƚ\u0003ĂƌĞ\u0003ŬŶŽǁŶ\u0003ƚŽ\u0003ďĞ\u0003ŵƵƚĂƟŽŶĂůůLJ\u0003ĂĐƟǀĞ.  \nKƵƌ\u0003ƐƉĂĐŝŶŐ\u0003ĂŶĚ\u0003ŽƌŝĞŶƚĂƟŽŶ\u0003ĂŶĂůLJƐĞƐ\u0003ĚĞŵŽŶƐƚƌĂƚĞ\u0003ƚŚĂƚ\u0003'Z,>Ϯ\u0003ďŝŶĚŝŶŐ\u0003ŝƐ\u0003ĂīĞĐƚĞĚ ďLJ\u0003ŵŽƟĨ\u0003\nŽƌŐ ĂŶŝnjĂƟŽŶ͘\u0003W ĂŝƌĞĚ\u0003ŵŽƟĨƐ\u0003ŝŶ\u0003Ă\u0003Ĩ ŽƌǁĂƌĚ–ĨŽƌǁĂƌĚ\u0003 ŽƌŝĞŶƚĂƟŽŶ\u0003(AACCGGTT-AACCGGTT) ĞdžŚŝďŝƚĞĚ\u0003\nƚŚĞ\u0003ƐƚƌŽŶŐĞƐƚ\u0003ďŝŶĚŝŶŐ\u0003Ăƚ\u0003ƐŚŽƌƚ\u0003ĚŝƐƚĂŶĐĞƐ\u0003ĂŶĚ\u0003ĚŝƐƉůĂLJĞĚ\u0003Ă\u0003ƉĞƌŝŽĚŝĐ\u0003ƉĂƩĞƌŶ\u0003ǁŝƚŚ\u0003ĂŶ\u0003ĂƉƉƌŽdžŝŵĂƚĞ\u0003\n5.5-bp interval, consistent with half of the DNA he lical turn (38). This periodicity suggests that \nŽƉƟŵĂů\u0003 ďŝŶĚŝŶŐ\u0003 ŽĐĐƵƌƐ\u0003 ǁŚĞŶ\u0003 ŵŽƟĨƐ\u0003 ĂƌĞ\u0003 ƉŽƐŝƟŽŶĞĚ\u0003 ŽŶ\u0003 ƚŚĞ\u0003 ƐĂŵĞ\u0003 ĨĂĐĞ\u0003 ŽĨ\u0003 ƚŚĞ\u0003 \u0018E\u0004\u0003 ŚĞůŝdž͕\u0003which \nsupports a ŵŽĚĞů\u0003ŽĨ\u0003ĐŽŽƉĞƌĂƟǀĞ\u0003Žƌ\u0003ĚŝŵĞƌ binding. While previous studies have suggested that \n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted April 18, 2026. ; https://doi.org/10.64898/2026.04.16.719077doi: bioRxiv preprint \n\n24 \n \nŵĞŵďĞƌƐ\u0003 ŽĨ\u0003 ƚŚĞ\u0003 'ƌĂŝŶLJŚĞĂĚ\u0003 ĨĂŵŝůLJ\u0003 ĐĂŶ\u0003 ĨŽƌŵ\u0003 ĚŝŵĞƌƐ͕\u0003 ƚŚĞ\u0003 ƉƌĞƐĞŶƚ\u0003 ƌĞƐƵůƚƐ\u0003 ĚĞĮŶĞ\u0003 ƚŚĞ\u0003 ƐƉĂƟĂů\u0003\nconstraints of ƚŚĞƐĞ\u0003 ŝŶƚĞƌĂĐƟŽŶƐ\u0003 ĂŶĚ\u0003 ƉƌŽǀŝĚĞ\u0003 Ă\u0003 ƐƚƌƵĐƚƵƌĂů\u0003 ďĂƐŝƐ\u0003 ĨŽƌ\u0003 ŵŽƟĨ\u0003 ĐůƵƐƚĞƌŝŶŐ\u0003 ŽďƐĞƌǀĞĚ\u0003\nwithin GRHL2-bound regulatory regions.  dŚĞ\u0003 ĚĞŶƐĞ\u0003 ŵŽƟĨ\u0003 ĐůƵƐƚĞƌ\u0003 ŽŶ\u0003 ĐŚƌŽŵŽƐŽŵĞ\u0003 ϳ͕\u0003 ǁŚĞƌĞ\u0003\nrepeated consensus and near-ĐŽŶƐĞŶƐƵƐ\u0003ŵŽƟĨƐ\u0003ƉƌŽĚƵĐĞĚ\u0003Ă\u0003ďƌŽĂĚ\u0003ƌĞŐŝŽŶ\u0003ŽĨ\u0003ĞůĞǀĂƚĞĚ\u0003^E\u0004W\u0003ƐŝŐŶĂů\u0003\nand ChIP-ƐĞƋ\u0003ŽĐĐƵƉĂŶĐLJ͕\u0003ƉƌŽǀŝĚĞƐ\u0003Ă\u0003ŶĂƚƵƌĂů\u0003ŐĞŶŽŵŝĐ\u0003ĞdžĂŵƉůĞ\u0003ŽĨ\u0003ŚŽǁ\u0003ƐƵĐŚ\u0003ŵƵůƟǀĂůĞŶƚ\u0003ďŝŶĚŝŶŐ\u0003\nenvironments may occur in the genome.  \n dŚĞƐĞ\u0003 ĮŶĚŝŶŐƐ\u0003 ƉƌŽǀŝĚĞ\u0003 Ă\u0003method ĨŽƌ\u0003 ŝŶƚĞƌƉƌĞƟŶŐ\u0003 'Z,>Ϯ\u0003 ĐŽ-occupancy with other \nƚƌĂŶƐĐƌŝƉƟŽŶ\u0003 ĨĂĐƚŽƌƐ\u0003 ŝŶ\u0003 ŐĞŶŽŵŝĐ\u0003 ƐƚƵĚŝĞƐ͘\u0003Recent work has demonstrated that GRHL2 co-binds \nĞŶŚĂŶĐĞƌ\u0003 ƌĞŐŝŽŶƐ\u0003 ǁŝƚŚ\u0003 WZ\u0003 ĂŶĚ\u0003 ĐŽŶƚƌŝďƵƚĞƐ\u0003 ƚŽ\u0003 ĐŚƌŽŵĂƟŶ\u0003 ĂĐĐĞƐƐŝďŝůŝƚLJ\u0003 ĂŶĚ\u0003 ƚƌĂŶƐĐƌŝƉƟŽŶĂů\u0003\nƌĞŐƵůĂƟŽŶ\u0003 ŝŶ\u0003 ŚŽƌŵŽŶĞ-responsive breast cancer cells (50). However, such approaches cannot \nresolve whether GRHL2 is directly engaging DNA at these sites or is recruited through protein-\nƉƌŽƚĞŝŶ\u0003ŝŶƚĞƌĂĐƟŽŶƐ͘\u0003\u0011LJ\u0003ĚŝƐƟŶŐƵŝƐŚŝŶŐ\u0003ĚŝƌĞĐƚ\u0003ŵŽƟĨ-driven binding from indirect occupancy, our \nresults act in complement with genome-wide studies to show that only a subset of GRHL2 ChIP-\nseq peaks represent sequence- ĞŶĐŽĚĞĚ\u0003 ďŝŶĚŝŶŐ\u0003 ĞǀĞŶƚƐ͕\u0003 ǁŚŝůĞ\u0003 ŽƚŚĞƌƐ\u0003 ŵĂLJ\u0003 ƌĞŇĞĐƚ\u0003 ĐŽĨĂĐƚŽƌ-\nŵĞĚŝĂƚĞĚ\u0003 ƌĞĐƌƵŝƚŵĞŶƚ͘\u0003 dŚŝƐ\u0003 ĚŝƐƟŶĐƟŽŶ\u0003 ŚĂƐ\u0003 ŝŵƉŽƌƚĂŶƚ\u0003 ŝŵƉůŝĐĂƟŽŶƐ\u0003ĨŽƌ\u0003 ŝŶƚĞƌƉƌĞƟŶŐ\u0003 ŐĞŶŽŵŝĐ\u0003\noccupancy data anĚ\u0003ĨŽƌ\u0003ŝĚĞŶƟĨLJŝŶŐ\u0003ĨƵŶĐƟŽŶĂůůLJ\u0003ƌĞůĞǀĂŶƚ\u0003'Z,>Ϯ\u0003ďŝŶĚŝŶŐ\u0003ƐŝƚĞƐ͘ dŚŝƐ\u0003ŽďƐĞƌǀĂƟŽŶ\u0003ŝƐ\u0003\nconsistent with prior transcriptomic and cistromic studies showing that GRHL2 interacts with ERɲ͕\u0003\nFŽdž\u0004ϭ͕\u0003ĂŶĚ\u0003ŽƚŚĞƌ\u0003ƚƌĂŶƐĐƌŝƉƟŽŶĂů\u0003ƌĞŐƵůĂƚŽƌƐ\u0003ĂŶĚ\u0003ĨƌĞƋƵĞŶƚůLJ\u0003ƉĂƌƟĐŝƉĂƚĞƐ\u0003ŝŶ\u0003ĞŶŚĂŶĐĞƌ\u0003ĞŶǀŝƌŽŶŵĞŶƚƐ\u0003\nƌĂƚŚĞƌ\u0003ƚŚĂŶ\u0003ĂĐƟŶŐ\u0003ĂƐ\u0003Ă\u0003solo DNA-binding factor (29, 30, 50, 51).  \n Towards ƉŽƚĞŶƟĂů\u0003ŝŵƉůŝĐĂƟŽŶƐ\u0003ŽĨ\u0003ƚŚĞƐĞ\u0003ĮŶĚŝŶŐƐ\u0003ĨŽƌ\u0003ďƌĞĂƐƚ\u0003ĐĂŶĐĞƌ\u0003ďŝŽůŽŐLJ, GRHL2 has been \nŝĚĞŶƟĮĞĚ ŝŶ\u0003 ŵĂŝŶƚĂŝŶŝŶŐ\u0003 ĞƉŝƚŚĞůŝĂů\u0003 ŝĚĞŶƟƚLJ\u0003 ĂŶĚ\u0003 ŝŶ\u0003 ŵŽĚƵůĂƟŶŐ\u0003 \u001cZɲ-ĚĞƉĞŶĚĞŶƚ\u0003 ƚƌĂŶƐĐƌŝƉƟŽŶĂů\u0003\nprograms, yet it has also been associated with worse clinical outcomes in ER-ƉŽƐŝƟǀĞ\u0003ƚƵŵŽƌƐ͘\u0003dŚĞ\u0003\n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted April 18, 2026. ; https://doi.org/10.64898/2026.04.16.719077doi: bioRxiv preprint \n\n25 \n \nĮŶĞ-scale binding ƐƉĞĐŝĮĐŝƚLJ\u0003 ƉĂƩĞƌŶƐ described here suggest that GRHL2 occupancy may be \nŚŝŐŚůLJ\u0003 ƐĞŶƐŝƟǀĞ\u0003 ƚŽ\u0003 ƐĞƋƵĞŶĐĞ\u0003 ƉĞƌƚƵƌďĂƟŽŶƐ\u0003 ǁŝƚŚŝŶ\u0003 ƚŚĞ\u0003central C dž dž ' \u0003Ž ƌ \u0003ƚ Ž \u0003Đ Ž Ž Ɖ Ğ ƌ Ă Ɵ ǀ Ğ \u0003Ɛ Ɖ Ă Đ ŝ Ŷ Ő \u0003\nƌĞůĂƟŽŶƐŚŝƉƐ\u0003Ăƚ\u0003ĚŝŵĞƌŝĐ\u0003sites͘\u0003^ŽŵĂƟĐ\u0003ŵƵƚĂƟŽŶƐ\u0003ĂīĞĐƟŶŐ\u0003ƚŚĞƐĞ\u0003ƉŽƐŝƟŽŶƐ or CpG instability at \nGRHL2-enriched regions ĐŽƵůĚ\u0003 ƐŚŝŌ\u0003 'Z,>Ϯ\u0003 ďŝŶĚŝŶŐ\u0003 ƉĂƩĞƌŶƐ\u0003 ŝŶ\u0003 ƚƵŵŽƌƐ͕\u0003 ƉŽƚĞŶƟĂůůLJ\u0003 ǁĞĂŬĞŶŝŶŐ\u0003\nepithelial maintenance programs or altering ER-ĂƐƐŽĐŝĂƚĞĚ\u0003ĞŶŚĂŶĐĞƌ\u0003ĂĐƟǀŝƚLJ͘\u0003\u0012ŽŶǀĞƌƐĞůLJ͕\u0003ƌĞŐŝŽŶƐ\u0003\nƚŚĂƚ\u0003ƌĞƚĂŝŶ\u0003ƐƚƌŽŶŐ\u0003'Z,>Ϯ\u0003ďŝŶĚŝŶŐ\u0003ĐĂƉĂĐŝƚLJ\u0003ŵĂLJ\u0003ƌĞŝŶĨŽƌĐĞ\u0003ĞƉŝƚŚĞůŝĂů\u0003ƚƌĂŶƐĐƌŝƉƟŽŶĂů\u0003ƐƚĂƚĞƐ\u0003Ănd could \nƌĞƉƌĞƐĞŶƚ\u0003 ĚŝƐƟŶĐƚ\u0003 ƚŚĞƌĂƉĞƵƟĐ\u0003 ƌĞƐƉŽŶƐĞ\u0003 ĐŽŶƚĞdžƚƐ͘\u0003 \u0004ůƚŚŽƵŐŚ\u0003 ŽƵƌ\u0003 ƐƚƵĚLJ\u0003 ĚŽĞƐ\u0003 ŶŽƚ\u0003 ĚŝƌĞĐƚůLJ\u0003 ĂƐƐĞƐƐ\u0003\nƚŚĞƐĞ\u0003 ƌĞůĂƟŽŶƐŚŝƉƐ͕\u0003 ƚŚĞ\u0003 ďŝŶĚŝŶŐ\u0003parameters established here provide ŝŶĨŽƌŵĂƟŽŶ towards \nŵĂƉƉŝŶŐ\u0003ŚŽǁ\u0003ƉĂƟĞŶƚ-ƐƉĞĐŝĮĐ\u0003ƐĞƋƵĞŶĐĞ\u0003ǀĂƌŝĂƟŽŶ\u0003ŵĂLJ\u0003ƌĞŵŽĚĞů\u0003ƚŚĞ\u0003'Z,>Ϯ\u0003ĐŝƐƚƌŽŵĞ͘ \n By analyzing GRHL2 binding at ChIP-seq sites in the absence of ĐŚƌŽŵĂƟŶ\u0003ĂŶĚ\u0003ĐŽŽƉĞƌĂƟǀĞ\u0003\nƚƌĂŶƐĐƌŝƉƟŽŶĂů\u0003 ĐŽ-regulators, this work provides a compendium of GRHL2 binding data for \n770,000-plus genomic sites and ĚĞĮŶĞƐ, at an unprecedented high ƌĞƐŽůƵƟŽŶ, the sequence \nƐƉĞĐŝĮĐŝƚLJ ƵŶĚĞƌůLJŝŶŐ\u0003 'Z,>Ϯ\u0003 \u0018E\u0004\u0003 ƌĞĐŽŐŶŝƟŽŶ\u0003 ĂĐƌŽƐƐ\u0003 ŐĞŶŽŵŝĐ\u0003 ƐĞƋƵĞŶĐĞ\u0003 ƐƉĂĐĞ͘\u0003The ability to \nresolve binding at sub-ƉĞĂŬ\u0003 ƌĞƐŽůƵƟŽŶ͕\u0003 ƋƵĂŶƟĨLJ\u0003 ƚŚĞ\u0003 ĞīĞĐƚƐ\u0003 ŽĨ\u0003 ƐĞƋƵĞŶĐĞ\u0003 ǀĂƌŝĂƟŽŶ͕\u0003 ĂŶĚ\u0003 ĚĞĮŶĞ\u0003\nspacing-ĚĞƉĞŶĚĞŶƚ\u0003ŝŶƚĞƌĂĐƟŽŶƐ\u0003ĂĚǀĂŶĐĞƐ\u0003ŽƵƌ\u0003ƵŶĚĞƌƐƚĂŶĚŝŶŐ ŽĨ\u0003'Z,>Ϯ\u0003ďĞLJŽŶĚ\u0003ĐŽŶƐĞŶƐƵƐ\u0003ŵŽƟĨ\u0003\nŵŽĚĞůƐ\u0003ƚŽǁĂƌĚ\u0003Ă\u0003ƉƌĞĚŝĐƟǀĞ͕\u0003ŵĞĐŚĂŶŝƐƟĐ\u0003ĚĞƐĐƌŝƉƟŽŶ\u0003ŽĨ\u0003ŝƚƐ\u0003\u0018E\u0004-ďŝŶĚŝŶŐ\u0003ƐƉĞĐŝĮĐŝƚLJ͘ KƵƌ\u0003ĮŶĚŝŶŐƐ\u0003\nshow that GRHL2 binding is anchored by a highly constrained central C dždžG within its octamer \nĐŽŶƐĞŶƐƵƐ\u0003ŵŽƟĨ\u0003and is ĂīĞĐƚĞĚ ďLJ\u0003ŶĞĂƌďLJ\u0003ŇĂŶŬŝŶŐ\u0003ƐĞƋƵĞŶĐĞ outside of its consensus. Our results \nsupport that GRHL2 ďŝŶĚŝŶŐ\u0003ŝƐ\u0003ƐƚƌĞŶŐƚŚĞŶĞĚ\u0003Ăƚ\u0003ĚŝŵĞƌŝĐ\u0003ƐŝƚĞƐ\u0003ĂŶĚ\u0003ĐĂŶ\u0003ďĞ\u0003ĚŝƐƟŶŐƵŝƐŚĞĚ\u0003ĂƐ\u0003ĞŝƚŚĞƌ\u0003\ndirect DNA binding or cofactor-mediated occupancy across the genome. dŚĞƐĞ\u0003ŝŶƐŝŐŚƚƐ\u0003ƌĞĮŶĞ\u0003the \nstructural and biochemical understanding of GRHL2 ŝŶƚĞƌĂĐƟŽŶ\u0003ǁŝƚŚ\u0003\u0018E\u0004 and lay the groundwork \nĨŽƌ\u0003 ĨƵƚƵƌĞ\u0003 ƐƚƵĚŝĞƐ\u0003 ůŝŶŬŝŶŐ\u0003 \u0012Ɖ'\u0003 ŵƵƚĂƟŽŶĂů\u0003 ůĂŶĚƐĐĂƉĞƐ with GRHL2- ĚĞƉĞŶĚĞŶƚ\u0003 ƚƌĂŶƐĐƌŝƉƟŽŶ\u0003 ŝŶ\u0003\n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted April 18, 2026. ; https://doi.org/10.64898/2026.04.16.719077doi: bioRxiv preprint \n\n26 \n \ndevelopment and disease, as well as methods to  disrupt key GRHL2 binding sites using site-\nĚŝƌĞĐƚĞĚ\u0003ƚŚĞƌĂƉĞƵƟĐƐ. \n\u0003\u0003\n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted April 18, 2026. ; https://doi.org/10.64898/2026.04.16.719077doi: bioRxiv preprint \n\n27 \n \n&/'hZ\u001c\u0003>\u001c'\u001cE\u0018^ \n&ŝŐƵƌĞ\u0003ϭ. &ƵŶĐƟŽŶĂů\u0003ĚŽŵĂŝŶ\u0003ŽƌŐĂŶŝnjĂƟŽŶ\u0003ŽĨ\u0003ŚƵŵĂŶ\u0003'Z,>Ϯ͘ Domain structure of GRHL2 (isoform \nϭͿ\u0003;\u0004ĐĐĞƐƐŝŽŶ\u0003ŶƵŵďĞƌ\u0003EWͺϬϳϵϭϵϭ͘ϮͿ ƐŚŽǁŝŶŐ\u0003ƚŚĞ\u0003ƚƌĂŶƐĂĐƟǀĂƟŽŶ\u0003ĚŽŵĂŝŶ͕\u0003\u0018E\u0004-binding domain \n;\u0018\u0011\u0018Ϳ͕\u0003ĐŽŶƐĞƌǀĞĚ\u0003\u0018\u0011\u0018\u0003ƐĞŐŵĞŶƚ͕\u0003ĂŶĚ\u0003ĚŝŵĞƌŝnjĂƟŽŶ\u0003ĚŽŵĂŝŶ, as adapted from (10). \n \n&ŝŐƵƌĞ\u0003 Ϯ.\u0003\u0018 Ğ Ɛ ŝ Ő Ŷ \u0003Ž Ĩ \u0003ƚ Ś Ğ \u0003^ E \u0004 W \u0003Ă ƌ ƌ Ă LJ \u0003Ğ Ŷ Ă ď ů Ğ Ɛ \u0003Ś ŝ Ő Ś-ƌĞƐŽůƵƟŽŶ\u0003 ŵĞĂƐƵƌĞŵĞŶƚ\u0003 ŽĨ\u0003 'Z,>Ϯ\u0003 ďŝŶĚŝŶŐ\u0003\nĂĐƌŽƐƐ\u0003 ŐĞŶŽŵŝĐ\u0003 \u0012Ś/W-ƐĞƋ\u0003 ƌĞŐŝŽŶƐ.\u0003\u0004 Ϳ \u0003>ĞŌ͗\u0003 WƵƌŝĮĞĚ\u0003 'Z,>Ϯ\u0003 ƉƌŽƚĞŝŶ\u0003 ĂŶĚ\u0003 ŇƵŽƌĞƐĐĞŶƚůLJ-labeled \n'Z,>Ϯ\u0003ĂŶƟďŽĚLJ\u0003ĂƐƐĂ LJ ĞĚ\u0003ŽŶ\u0003ƚŚĞ\u0003^E\u0004W\u0003Ăƌƌ Ă LJ ͘\u0003DŝĚĚůĞ͗\u0003^ƵďƐĞĐƟŽŶ\u0003ŽĨ\u0003ƚŚĞ\u0003'Z,>Ϯ\u0003ĂƌƌĂLJ\u0003ĚŝƐƉůĂLJŝŶŐ\u0003\n'Z,>Ϯ\u0003ǁŝƚŚ\u0003ĂŶƟďŽĚLJ\u0003ďŽƵŶĚ\u0003ƚŽ\u0003\u0018E\u0004\u0003ƉŽůLJŵĞƌĂƐĞ\u0003ĞdžƚĞŶĚĞĚ double-stranded probes. Right: Probes \nconsisted of variable 48-nt genomic regions (green) with a 12-nt constant region (black) on the 3’ \nend ƵƐĞĚ\u0003ĨŽƌ\u0003ƉƌŝŵĞƌ\u0003ĞdžƚĞŶƐŝŽŶ͘\u0003dŚĞ\u0003microarray strand ŝƐ\u0003ĂƩĂĐŚĞĚ\u0003ƚŽ\u0003ƚŚĞ\u0003^E\u0004W\u0003ĂƌƌĂLJ\u0003ƐƵƌĨĂĐĞ\u0003Ăƚ\u0003ƚŚĞ\u0003\n3' terminus. \u0011Ϳ\u0003Peak overlap of the four GRHL2 ChIP-seq datasets used to construct the genomic \nƟůŝŶŐ\u0003ƌĞŐŝŽŶƐ\u0003ŽĨ\u0003ƚŚĞ\u0003^E\u0004W\u0003ĂƌƌĂLJ͘ \u0012Ϳ\u0003Close-up of a region of the full SNAP array to show varying \nŝŶƚĞŶƐŝƟĞƐ\u0003 ŽĨ\u0003 ŝŶĚŝǀŝĚƵĂů\u0003 ďŽƵŶĚ\u0003 ĨĞĂƚƵƌĞƐ. GRHL2 binding is detected as red circles. DͿ Top: \nGRHL2 ChIP-seq peak from Reese et al. (30) with the ƉŽƐŝƟŽŶ\u0003 ŽĨ the OVOL1 gene \ndisplayed. \u0011 Ž Ʃ Ž ŵ ͗ \u0003' Ğ Ŷ Ž ŵ ŝ Đ \u0003^ E \u0004 W \u0003Ɖ ƌ Ž ď Ğ Ɛ \u0003Ɵ ů ŝ Ŷ Ő \u0003ƚ Ś Ğ ChIP-seq peak at 6-bp spacing with the \nŶŽƌŵĂůŝnjĞĚ\u0003 ^E\u0004W\u0003 ŝŶƚĞŶƐŝƚLJ\u0003 ĨŽƌ\u0003 ĞĂĐŚ\u0003 ƉƌŽďĞ\u0003 ĚŝƐƉůĂLJĞĚ\u0003 ƚŽ\u0003 ƚŚĞ\u0003 ƌŝŐŚƚ͘\u0003 dŚĞ\u0003 ĐLJĂŶ\u0003 ďŽdž\u0003 ĚĞŶŽƚĞƐ\u0003 Ă\u0003\nconsensus GRHL2 site on the SNAP probe. The green bases in the SNAP probes correspond to \ngenomic sequence while the black bases correspond to the constant primer.  \n \n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted April 18, 2026. ; https://doi.org/10.64898/2026.04.16.719077doi: bioRxiv preprint \n\n28 \n \n&ŝŐƵƌĞ\u0003ϯ͘\u0003^E\u0004W\u0003ŐĞŶŽŵŝĐ\u0003ďŝŶĚŝŶŐ\u0003ĚĂƚĂ\u0003ƌĞĐŽǀĞƌƐ\u0003ƚŚĞ\u0003'Z,>Ϯ\u0003ĐŽŶƐĞŶƐƵƐ\u0003ŵŽƟĨ\u0003ĂŶĚ\u0003ƌĞǀĞĂůƐ\u0003ŇĂŶŬŝŶŐ\u0003\nƐĞƋƵĞŶĐĞ\u0003 ƉƌĞĨĞƌĞŶĐĞƐ\u0003 ĂŶĚ\u0003 ŵŝƐŵĂƚĐŚ\u0003 ĚŝƐƚƌŝďƵƟŽŶs. \u0004Ϳ \u001cdžƉĞĐƚĂƟŽŶ\u0003 ŵĂdžŝŵŝnjĂƟŽŶ\u0003 ĂůŐŽƌŝƚŚŵ\u0003\nŵŽƟĨƐ\u0003ĨŽƌ\u0003ƚŚĞ\u0003ŚŝŐŚĞƐƚ\u0003ŝŶƚĞŶƐŝƚLJ\u0003^E\u0004W\u0003ŐĞŶŽŵŝĐ\u0003ƐĞƋƵĞŶĐĞƐ\u0003ǁŝƚŚ\u0003Ă\u0003ĚĞĮŶĞĚ\u0003ŵŽƟĨ\u0003ǁŝĚƚŚ\u0003ŽĨ\u0003ϭϬ\u0003ďĂƐĞ\u0003\npairs (bp). DŽƟĨ\u0003ĨƌŽŵ\u0003ƚŚĞ\u0003ƚŽƉ\u0003ϯϬϬ\u0003;ƚŽƉ\u0003ůĞŌͿ͕\u0003ϯϬϭ-600 (top right), 601-ϵϬϬ\u0003;ďŽƩŽŵ\u0003ůĞŌͿ͕\u0003ĂŶĚ\u0003ϵϬϭ-\nϭϮϬϬ\u0003;ďŽƩŽŵ\u0003right) highest intensity SNAP genomic sequences. WŽƐŝƟŽŶ\u0003ŶƵŵďĞƌŝŶŐ\u0003ŝƐ\u0003ďĂƐĞĚ\u0003ŽŶ\u0003\nthe start of the GRHL2 octamer consensus, 5’-AACCGGTT-3’. \u0011Ϳ DŽƟĨ\u0003ŽĨ\u0003ƚŚĞ\u0003ϯϬϬ\u0003ŚŝŐŚĞƐƚ\u0003ŝŶƚĞŶƐŝƚLJ\u0003\nSNAP genomic sequences containing ĞdžĂĐƚůLJ\u0003ŽŶĞ\u0003ŵŝƐŵĂƚĐŚ\u0003ŝŶ\u0003ƚŚĞ\u0003'Z,>Ϯ\u0003ĐŽŶƐĞŶƐƵƐ\u0003ŵŽƟĨ͘\u0003&ŝǀĞ\u0003\nŇĂŶŬŝŶŐ\u0003ďĂƐĞƐ\u0003ĂƌĞ\u0003ŝŶĐůƵĚĞĚ\u0003ŝŶ\u0003ƚŚĞ\u0003ĚŝƐƉůĂLJĞĚ\u0003ŵŽƟĨ͘\u0003\u0012Ϳ\u0003\u0018ŝƐĐƌŝŵŝŶĂƚŽƌLJ\u0003ĂŶĂůLJƐŝƐ\u0003ŽĨ\u0003ŇĂŶŬŝŶŐ\u0003ƐĞƋƵĞŶĐĞ\u0003\nsurrounding the GRHL2 consensus using the highest versus lowest intensity genomic sequences. \ndŚŝƐ\u0003 ƐĞƋƵĞŶĐĞ\u0003 ƐĞƚ\u0003 ŝƐ\u0003 ĐŽŵƉŽƐĞĚ\u0003 ĞdžĐůƵƐŝǀĞůLJ\u0003 ŽĨ\u0003 ŐĞŶŽŵŝĐ\u0003 ƉƌŽďĞƐ\u0003 ƚŚĂƚ\u0003 ĐŽŶƚĂŝŶ\u0003 ĂŶ\u0003 ĞdžĂĐƚ\u0003 ĐĂŶŽŶŝĐĂů\u0003\nGRHL2 consensus octamer, unlike ;\u0004-\u0011Ϳ. Top: A summary sequence of the consensus (black) and \nĂŶĂůLJnjĞĚ\u0003 ŇĂŶŬŝŶŐ\u0003 ƐĞƋƵĞŶĐĞ\u0003 ;ŐƌĞĞŶͿ\u0003 ŝƐ\u0003 ƉƌŽǀŝĚĞĚ͕\u0003 ǁŚĞƌĞ\u0003 ƵŶĚĞƌƐĐŽƌĞƐ\u0003 ŝŶĚŝĐĂƚĞ\u0003 ŶŽ\u0003 ƐŝŐŶŝĮĐĂnt \nĚŝƐĐƌŝŵŝŶĂƚŽƌLJ\u0003ƉƌĞĨĞƌĞŶĐĞ͘\u0003\u0011ŽƩŽŵ͗\u0003dĂďůĞ\u0003ůŝƐƟŶŐ\u0003ĞŶƌŝĐŚĞĚ\u0003ďĂƐĞƐ\u0003Ăƚ\u0003ƐƉĞĐŝĮĐ\u0003ŇĂŶŬŝŶŐ\u0003ƉŽƐŝƟŽŶƐ͘\u0003dŚĞ\u0003\nŶƵŵďĞƌ\u0003ŽĨ\u0003ĂƐƚĞƌŝƐŬƐ\u0003ĚĞŶŽƚĞ\u0003ƚŚĞ\u0003ƐŝŐŶŝĮĐĂŶĐĞ\u0003ŽĨ\u0003ƚŚĞ\u0003Ɖ-value. \u0018Ϳ SSL of all SNAP genomic sequences \nƵƐŝŶŐ\u0003ƚŚĞ\u0003ĐŽŶƐĞŶƐƵƐ\u0003'Z,>Ϯ\u0003ŵŽƟĨ\u0003;ϱΖ-AACCGGTT-ϯΖͿ\u0003ĂƐ\u0003ƚŚĞ\u0003ĐĞŶƚƌĂů\u0003^^>\u0003ŵŽƟĨ͘\u0003\u001cͿ Two-dimensional \nƌĞƉƌĞƐĞŶƚĂƟŽŶ\u0003 ŽĨ\u0003 ƚŚĞ\u0003 ^^>\u0003 ϭ-mismatch ring. This ring is divided into regions corresponding to \nwhere the mismatch to the GRHL2 consensus occurs. Genomically, mismatches in the central two \nƉŽƐŝƟŽŶƐ\u0003ŽĨ\u0003ƚŚĞ\u0003ŵŽƟĨ\u0003ŽĐĐƵƌ\u0003ŵƵch more commonly (~10- to 20-ĨŽůĚͿ\u0003ƚŚĂŶ\u0003ƚŚĞ\u0003ŽƚŚĞƌ\u0003ƉŽƐŝƟŽŶƐ͘\u0003dŚĞ\u0003\nŶ Ƶ ŵ ď Ğ ƌ \u0003Ž Ĩ \u0003Ɖ ƌ Ž ď Ğ Ɛ \u0003Ă ƚ \u0003ƚ Ś Ğ \u0003Ě ŝ ī Ğ ƌ ŝ Ŷ Ő \u0003ŵ ŝ Ɛ ŵ Ă ƚ Đ Ś \u0003Ɖ Ž Ɛ ŝ Ɵ Ž Ŷ Ɛ \u0003ŝ Ɛ \u0003Ɛ LJ ŵ ŵ Ğ ƚ ƌ ŝ ĐĂ ů \u0003Ɛ ŝ Ŷ Đ Ğ \u0003ƚ Ś Ğ \u0003ŵ Ž Ɵ Ĩ \u0003ŝ Ɛ \u0003\npalindromic. \n \n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted April 18, 2026. ; https://doi.org/10.64898/2026.04.16.719077doi: bioRxiv preprint \n\n29 \n \n&ŝŐƵƌĞ\u0003ϰ.\u0003^LJƐƚĞŵĂƟĐ\u0003ŵŝƐŵĂƚĐŚ\u0003ĂŶĂůLJƐŝƐ\u0003ƌĞǀĞĂůƐ\u0003ƐĞƋƵĞŶĐĞ\u0003ƌƵůĞƐ\u0003ŐŽǀĞƌŶŝŶŐ\u0003'Z,>Ϯ\u0003ŵŽŶŽŵĞƌ\u0003ĂŶĚ\u0003\nĚŝŵĞƌ\u0003ďŝŶĚŝŶŐ. \u0004Ϳ \u001cdžƉĞĐƚĂƟŽŶ\u0003ŵĂdžŝŵŝnjĂƟŽŶ\u0003ĂůŐŽƌŝƚŚŵ\u0003ŵŽƟĨ\u0003ĨŽƌ\u0003ƚŚĞ\u0003ϭϮϬϬ\u0003ŚŝŐŚĞƐƚ\u0003ŝŶƚĞŶƐŝƚLJ\u0003^E\u0004W\u0003\nŐĞŶŽŵŝĐ\u0003ƐĞƋƵĞŶĐĞƐ\u0003ǁŝƚŚ\u0003Ă\u0003ĚĞĮŶĞĚ\u0003ŵŽƟĨ\u0003ǁŝĚƚŚ\u0003ŽĨ\u0003ϭϴ\u0003ďĂƐĞ\u0003ƉĂŝƌƐ\u0003;ďƉͿ͘\u0003\u0011Ϳ\u0003Crystal structure of GRHL2 \nDNA-binding domain (PDB͗\u0003ϱDZϳͿ\u0003ǁŝƚŚ\u0003ŽǀĞƌůĂŝĚ\u0003ůŝŶĞĂƌ\u0003ĚƵƉůĞdž\u0003\u0018E\u0004\u0003ƐĐŚĞŵĂƟĐ\u0003ƉůĂĐĞĚ\u0003ŽŶ\u0003ƚŚĞ\u0003ďĂƐŝƐ\u0003\nof the GRHL1-DNA crystal structure (PDB: 5MPF). \u0012Ϳ\u0003Average normalized SNAP intensity for an \nincreasing number of mismatches to the consensus sequence 5’-AACCGGTT-3’. Each mismatch \ngroup represents all possible mismatches to the consensus͕\u0003ƌĞƐƵůƟŶŐ\u0003ŝŶ\u0003Ϯϰ\u0003ƵŶŝƋƵĞ\u0003ŽŶĞ-mismatch \nprobes, 252 unique two-mismatch probes, 1512 unique three-mismatch probes, and 5670 unique \nfour-mismatch probes, placed on the SNAP array in triplicate͘\u0003dŚĞ\u0003ŵŽƟĨƐ\u0003ĂƌĞ\u0003ĂŶĐŚŽƌĞĚ\u0003ŽŶ\u0003ĞŝƚŚĞƌ\u0003\nside by constant regions (C1 and C2). Error bars ;ƌĞƉƌĞƐĞŶƟŶŐ\u0003ŽŶĞ\u0003ƐƚĂŶĚĂƌĚ\u0003ĚĞǀŝĂƟŽŶͿ\u0003are derived \nĨƌŽŵ\u0003 Ăůů\u0003 ŝŶƚĞŶƐŝƟĞƐ\u0003 ŽĨ\u0003 Ăůů\u0003 ƌĞƉůŝĐĂƚĞ\u0003 ƉƌŽďĞƐ\u0003 ĂŶĚ\u0003 ƉŽƐƐŝďůĞ\u0003 ŵŝƐŵĂƚĐŚ\u0003 ǀĂƌŝĂƟŽŶƐ\u0003 ǁŝƚŚŝŶ\u0003 ƚŚĞ\u0003\ngroup.  \u0018Ϳ \u001cīĞĐƚƐ\u0003ŽĨ\u0003ŵŝƐŵĂƚĐŚ(es) in a dimeric GRHL2 consensus site separated by a single base \nƉĂŝƌ͕\u0003ŇĂŶŬĞĚ\u0003ďLJ\u0003ĐŽŶƐƚĂŶƚ\u0003ƌĞŐŝŽŶƐ͘\u0003dŚĞ\u0003dž-ĂdžŝƐ\u0003ůĂďĞůƐ\u0003ĚĞŶŽƚĞ\u0003ƚŚĞ\u0003ŶƵŵďĞƌ\u0003ŽĨ\u0003ŵŝƐŵĂƚĐŚĞƐ\u0003ŝŶ\u0003ƚŚĞ\u0003ůĞŌ\u0003\nbinding site followed by the number of mismatch(e s) in the right binding site, separated by an \nunĚĞƌƐĐŽƌĞ͘\u0003\u001cĂĐŚ\u0003ƉƌŽďĞ\u0003ƌĞƉƌĞƐĞŶƚĞĚ\u0003ŝŶ\u0003ƚŚĞ\u0003ĮƌƐƚ\u0003ĮǀĞ ĐŽůƵŵŶƐ\u0003ĐŽŶƐŝƐƚƐ\u0003ŽĨ\u0003ŝĚĞŶƟĐĂů mismatches in \nthe two binding sites, while the middle and last four columns consist of mismatch(es) only in the \nĮƌƐƚ\u0003 Žƌ\u0003 ƐĞĐŽŶĚ\u0003 ďŝŶĚŝŶŐ\u0003 ƐŝƚĞ͕\u0003 ƌĞƐƉĞĐƟǀĞůLJ͕\u0003 ǁŝƚŚ\u0003 ƚŚĞ\u0003 ŽƚŚĞƌ\u0003 ďŝŶĚŝŶŐ \u0003Ɛ ŝ ƚ Ğ \u0003Đ Ž ŵ Ɖ ƌ ŝ Ɛ Ğ Ě \u0003Ž Ĩ \u0003ƚ Ś Ğ \u0003\nconsensus. \u001cͿ \u0004ǀĞƌĂŐĞ\u0003 ŶŽƌŵĂůŝnjĞĚ\u0003 ^E\u0004W\u0003 ŝŶƚĞŶƐŝƟĞƐ\u0003 ĨŽƌ\u0003 ĞĂĐŚ\u0003 ƐŝŶŐůĞ\u0003 ŵŝƐŵĂƚĐŚ\u0003 ƚŽ\u0003ƚŚĞ\u0003 'Z,>Ϯ\u0003\nĐŽŶƐĞŶƐƵƐ\u0003ƐŝƚĞ\u0003ďĂƐĞĚ\u0003ŽŶ\u0003ĐŽŶƚƌŽů\u0003ƉƌŽďĞ\u0003ŝŶƚĞŶƐŝƟĞƐ͘ Red bases denote mismatches to the GRHL2 \nconsensus. &Ϳ \u0004ǀĞƌĂŐĞ\u0003ŶŽƌŵĂůŝnjĞĚ\u0003^E\u0004W\u0003ŝŶƚĞŶƐŝƟĞƐ\u0003ĨŽƌ\u0003ƚŚĞ\u0003Ɛŝdž highest and seven lowest intensity \nŵŝƐŵĂƚĐŚ\u0003ŵŽƟĨƐ\u0003ǁŚĞƌĞ\u0003ƚŚĞ\u0003ŶƵŵďĞƌ\u0003ŽĨ\u0003ŵŝƐŵĂƚĐŚĞƐ to the canonical GR,>Ϯ\u0003ŵŽƟĨ vary up to four. \nRed bases denote mismatches to the GRHL2 consensus. \n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted April 18, 2026. ; https://doi.org/10.64898/2026.04.16.719077doi: bioRxiv preprint \n\n30 \n \n&ŝŐƵƌĞ\u0003 ϱ͘\u0003 'Z,>Ϯ\u0003 ĚŝŵĞƌ\u0003 ďŝŶĚŝŶŐ\u0003 ƐƚƌĞŶŐƚŚ\u0003 ŽƐĐŝůůĂƚĞƐ\u0003 ǁŝƚŚ\u0003 \u0018E\u0004\u0003 ŚĞůŝĐĂů\u0003 ƉĞƌŝŽĚŝĐŝƚLJ͘\u0003 \u0004Ϳ Mean \nŇƵŽƌĞƐĐĞŶĐĞ\u0003ŝŶƚĞŶƐŝƚLJ\u0003;ц\u0003^\u001cDͿ\u0003ĨŽƌ\u0003ĨĞĂƚƵƌĞƐ\u0003ĐŽŶƚĂŝŶŝŶŐ\u0003ƚǁŽ\u0003'Z,>Ϯ\u0003ŵŽƟĨƐ\u0003ƐĞƉĂƌĂƚĞĚ\u0003ďLJ\u0003Ϭ-32 bp in \nforward-forward (gray) or reverse-forward (whiteͿ\u0003ŽƌŝĞŶƚĂƟŽŶ͕\u0003ŝůůƵƐƚƌĂƟŶŐ\u0003ŽƌŝĞŶƚĂƟŽŶ-dependent \nĚŝīĞƌĞŶĐĞƐ\u0003 ŝŶ\u0003 ďŝŶĚŝŶŐ\u0003 ƐŝŐŶĂů͘\u0003\u0011Ϳ Forward-ĨŽƌǁĂƌĚ\u0003 ŇƵŽƌĞƐĐĞŶĐĞ\u0003 ǀĂůƵĞƐ\u0003 ƉůŽƩĞĚ\u0003 ŽŶ\u0003 Ă\u0003 ůŽŐĂƌŝƚŚŵŝĐ\u0003\nƐ Đ ĂůĞ\u0003ĂƐ\u0003Ă\u0003Ĩ ƵŶĐƟ Ž Ŷ\u0003Ž Ĩ\u0003ĐĞŶ ƚ Ğƌ-to-ĐĞŶƚĞƌ\u0003ŵŽƟĨ\u0003ƐƉĂĐŝŶŐ͘\u0003WŽŝŶƚƐ\u0003ƌĞƉƌĞƐĞŶƚ ŵĞĂŶ\u0003ц\u0003^\u001cD͘\u0003dŚĞ\u0003ďůĂĐŬ\u0003\ncurve shows the best-Įƚ\u0003ƚŽƌƐŝŽŶĂů\u0003ƉŚĂƐŝŶŐ\u0003ŵŽĚĞů͕\u0003LJŝĞůĚŝŶŐ\u0003ƉĞƌŝŽĚŝĐŝƚLJ\u0003ŽĨ\u0003ĂƉƉƌŽdžŝŵĂƚĞůLJ\u0003ϱ͘ϰϲ\u0003ďƉ͘\u0003\n\u0018 Ă ƚ Ă \u0003Ɖ Ž ŝ Ŷ ƚ Ɛ \u0003Ă ƌ Ğ \u0003Đ Ž ů Ž ƌ Ğ Ě \u0003ď LJ \u0003ƚ Ś Ğ \u0003Ɖ Ś Ă Ɛ Ğ \u0003Ž Ĩ \u0003ƚ Ś Ğ \u0003Į Ʃ Ğ Ě \u0003Ś Ğ ů ŝ Đ Ă ů \u0003Ɖ Ğ ƌ ŝ Ž ĚŝĐŝƚLJ\u0003 ŵŽĚĞů͘\u0003\u0012Ϳ Helical \nƌĞƉƌĞƐĞŶƚĂƟŽŶ\u0003ŽĨ\u0003ƚŚĞ\u0003ĨŽƌǁĂƌĚ-ĨŽƌǁĂƌĚ\u0003ĚĂƚĂ͕\u0003ŵĂƉƉŝŶŐ\u0003ŵŽƟĨ\u0003ƐƉĂĐŝŶŐ\u0003ŽŶƚŽ\u0003ƚŚĞ\u0003ĞdžƉĞĐƚĞĚ\u0003ŐĞŽŵĞƚƌLJ\u0003\nŽĨ\u0003ƚŚĞ\u0003\u0018E\u0004\u0003ŚĞůŝdž\u0003;ϭϬ͘ϱ\u0003ďƉ\u0003ƉĞƌ\u0003ƚƵƌŶͿ͘\u0003WŽŝŶƚƐ\u0003ƌĞƚĂŝŶ\u0003ĐŽůŽƌ\u0003ĐŽĚŝŶŐ\u0003ĨƌŽŵ\u0003ƉĂŶĞů\u0003(\u0011Ϳ͕\u0003ŚŝŐŚůŝŐŚƟŶŐ\u0003ƚŚĂƚ\u0003\nƉĞĂŬ\u0003ďŝŶĚŝŶŐ\u0003ŝŶƚĞŶƐŝƟĞƐ\u0003ĐůƵƐƚĞƌ\u0003ŽŶ\u0003ĂůƚĞƌŶĂƟŶŐ\u0003ĨĂĐĞƐ\u0003ŽĨ\u0003ƚŚĞ\u0003\u0018E\u0004\u0003ŚĞůŝdž͕\u0003ĐŽŶƐŝƐƚĞŶƚ\u0003ǁŝƚŚ\u0003ƉƌĞĨĞƌĞŶƟĂů\u0003\nbinding when GRHL2 dimers occupy the same helical face. \u0018Ϳ Structural model ŝůůƵƐƚƌĂƟŶŐ\u0003ƚŚĞ\u0003\nƉƌĞĚŝĐƚĞĚ\u0003 ĐŽŶĮŐƵƌĂƟŽŶ\u0003 Ăƚ\u0003 ƚŚĞ\u0003 ŽƉƟŵĂů\u0003 ĨŽƌǁĂƌĚ-forward spacing of 1 bp. Using the crystal \nstructure of GRHL1 dimer bound to DNA, two GRHL2 dimers, depicted as yellow and salmon \nmonomers forming dimer 1 and white and blue monomers forming dimer 2, ǁĞƌĞ\u0003ƉŽƐŝƟŽŶĞĚ\u0003ƚŽ\u0003\nŵŽĚĞů\u0003ďŝŶĚŝŶŐ\u0003ƚŽ\u0003ƚǁŽ\u0003\u0018E\u0004\u0003ŵŽƟĨƐ\u0003ƐƉĂĐĞĚ\u0003ϭ\u0003ďƉ\u0003ĂƉĂƌƚ͘\u0003\n\u0003\n&ŝŐƵƌĞ\u0003ϲ͘\u0003\u0003^E\u0004W\u0003ĚĂƚĂ\u0003ĚŝƐƟŶŐƵŝƐŚĞĚ\u0003ĚŝƌĞĐƚ\u0003'Z,>Ϯ\u0003\u0018E\u0004\u0003ďŝŶĚŝŶŐ\u0003ĨƌŽŵ\u0003ŝŶĚŝƌĞĐƚ\u0003ƉƌŽƚĞŝŶ-ŵĞĚŝĂƚĞĚ\u0003\nƌĞĐƌƵŝƚŵĞŶƚ\u0003 Ăƚ\u0003 ŐĞŶŽŵŝĐ\u0003 ƐŝƚĞƐ. \u0004Ϳ ^ĐŚĞŵĂƟĐ\u0003 ĚĞŵŽŶƐƚƌĂƟŶŐ\u0003 ƚŚĞ\u0003 ŵĞƚŚŽĚ\u0003 ƚŽ\u0003 ĚĞƚĞƌŵŝŶĞ\u0003 ĚŝƌĞĐƚ\u0003\nǀĞƌƐƵƐ\u0003ŝŶĚŝƌĞĐƚ\u0003ďŝŶĚŝŶŐ\u0003ŽĨ\u0003'Z,>Ϯ\u0003Ăƚ\u0003ŐĞŶŽŵŝĐ\u0003ƐŝƚĞƐ͘\u0003>ĞŌ͗\u0003tŚĞŶ\u0003ďŝŶĚŝŶŐ\u0003ŝƐ\u0003ŽďƐĞƌǀĞĚ\u0003ďLJ\u0003ďŽƚŚ\u0003\u0012Ś/W-\nƐĞƋ\u0003ĂŶĚ\u0003^E\u0004W\u0003ĂƌƌĂLJƐ\u0003Ăƚ\u0003ƚŚĞ\u0003ƐĂŵĞ\u0003ŐĞŶŽŵŝĐ\u0003ůŽĐĂƟŽŶ͕\u0003ƚŚŝƐ\u0003ŝƐ\u0003ĐŽŶƐŝĚĞƌĞĚ\u0003Ă\u0003ĚŝƌĞĐƚ\u0003ďŝŶĚŝŶŐ\u0003ĞǀĞŶƚ͘\u0003ZŝŐŚƚ͗\u0003\nWhen binding is observed by ChIP-seq but not SNAP arrays, then the ChIP-seq binding is \nconsidered to be mediated by a secondary cofactor (CoF) that indirectly tethers GRHL2 to this \n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted April 18, 2026. ; https://doi.org/10.64898/2026.04.16.719077doi: bioRxiv preprint \n\n31 \n \nŐĞŶŽŵŝĐ\u0003ůŽĐĂƟŽŶ͘\u0003\u0011Ϳ\u0003ZĞƉƌĞƐĞŶƚĂƟǀĞ\u0003ĞdžĂŵƉůĞƐ\u0003ŽĨ\u0003Ă\u0003ĚŝƌĞĐƚ\u0003;>ĞŌͿ\u0003ĂŶĚ\u0003ŝŶĚŝƌĞĐƚ\u0003;ZŝŐŚƚͿ\u0003ďŝŶĚŝŶŐ\u0003ĞǀĞŶƚ͘\u0003\nChIP-ƐĞƋ\u0003ƐŝŐŶĂů\u0003ĨƌŽŵ\u0003ŽǀĞƌĞdžƉƌĞƐƐĞĚ\u0003;dŽƉͿ\u0003ĂŶĚ\u0003ĞŶĚŽŐĞŶŽƵƐ\u0003;DŝĚĚůĞͿ\u0003'Z,>Ϯ\u0003ůĞǀĞůƐ\u0003Ăƚ\u0003ƚŚĞ\u0003D\u0012d^ϭ\u0003\nand EPIC1 promoters (7, 30) ͘\u0003 EŽƌŵĂůŝnjĞĚ\u0003 'Z,>Ϯ\u0003 ďŝŶĚŝŶŐ\u0003 ŝŶƚĞŶƐŝƚLJ\u0003 ;\u0011ŽƩŽŵͿ\u0003is shown at these \npromoters ĂƐ\u0003ƟůĞĚ\u0003ŽŶ\u0003ƚŚĞ\u0003^E\u0004W\u0003ĂƌƌĂLJ͘\u0003\u001cĂĐŚ\u0003ƌĞŐŝŽŶ\u0003ĐŽŶƚĂŝŶƐ\u0003ŵƵůƟƉůĞ\u0003\u0018E\u0004\u0003ƉƌŽďĞƐ\u0003ƟůĞĚ\u0003ĂĐƌŽƐƐ\u0003ƚŚĞ\u0003\nregion. \u0012Ϳ\u0003Summary of the genomic probe count on the SNAP array with the corresponding \nnumber of genomic regions evaluated using overlapping SNAP probes. These genomic regions are \nthen divided into direct and indirect GRHL2 site s based on the SNAP probe in each region with \nthe highest intensity.  \u0018Ϳ \u0011Ăƌ\u0003 ŐƌĂƉŚ\u0003 ŽĨ\u0003 ƚŚĞ\u0003 ŶƵŵďĞƌ\u0003 ŽĨ\u0003 ĚŝƐƟŶĐƚ\u0003 'Z,>Ϯ\u0003 ďŝŶĚŝŶŐ\u0003 Ɛites within each \nŐĞŶŽŵŝĐ\u0003ƌĞŐŝŽŶ\u0003ƟůĞĚ\u0003ďLJ\u0003ƚŚĞ\u0003^E\u0004W\u0003ĂƌƌĂLJ͘\u0003KŶĞ\u0003ƌĞŐŝŽŶ͕\u0003ĐŚƌϳ͗ϭϱϱ͕ϭϮϯ͕ϰϬϬ-155,125,400 (Figure 7), \nǁĂƐ\u0003ĞdžĐůƵĚĞĚ\u0003ĨƌŽŵ\u0003ƚŚŝƐ\u0003ĂŶĂůLJƐŝƐ\u0003ĚƵĞ\u0003Ă\u0003ůĂƌŐĞ\u0003ŶƵŵďĞƌ\u0003ŽĨ\u0003ĚŝƌĞĐƚ\u0003ďŝŶĚŝŶŐ\u0003ƐŝƚĞƐ\u0003ǁŝƚŚŝŶ\u0003Ă\u0003ǀĞƌLJ\u0003ƐŚŽƌƚ\u0003\nregion, where numerous SNAP ƉƌŽďĞƐ\u0003ĐŽŶƚĂŝŶ\u0003ŵƵůƟƉůĞ\u0003'Z,>Ϯ\u0003ĚŝƌĞĐƚ\u0003ďŝŶĚŝŶŐ\u0003ƐŝƚĞƐ͘ \n \n&ŝŐƵƌĞ\u0003ϳ͘\u0003\u0004\u0003ŐĞŶŽŵŝĐ\u0003ƌĞŐŝŽŶ\u0003ǁŝƚŚ\u0003ĞdžƚƌĞŵĞ\u0003'Z,>Ϯ\u0003ŵŽƟĨ\u0003ĚĞŶƐŝƚLJ\u0003ƉƌŽĚƵĐĞƐ\u0003ďƌŽĂĚ\u0003^E\u0004W\u0003ďŝŶĚŝŶŐ\u0003\nƐŝŐŶĂů\u0003ƚŚĂƚ\u0003ŝƐ\u0003ŶŽƚ\u0003ĨƵůůLJ\u0003ƌĞŇĞĐƚĞĚ\u0003ŝŶ\u0003\u0012Ś/W-ƐĞƋ\u0003ƉƌŽĮůĞƐ. \u0004Ϳ Genomic sequence of chr7:155,124,004-\nϭϱϱ͕ϭϮϰ͕ϳϱϵ\u0003 ;'Z\u0012ŚϯϴͿ͘\u0003 \u0012ŽŶƐĞŶƐƵƐ\u0003 ŵŽƟĨƐ\u0003 ĂƌĞ\u0003 ƵŶĚĞƌůŝŶĞĚ\u0003 ǁŝƚŚ\u0003 ƉĞƌĨĞĐƚ\u0003 ƐŝƚĞƐ\u0003 ŝŶ\u0003 ƌĞĚ\u0003 ĂŶĚ\u0003 ƐŝŶŐůĞ\u0003\nŵŝƐŵĂƚĐŚ\u0003ƐŝƚĞƐ\u0003ŝŶ\u0003ďůƵĞ͘\u0003>Žǁ\u0003ĐŽŵƉůĞdžŝƚLJ\u0003ƌĞƉĞĂƚƐ\u0003ŽĨ\u0003ϮϮ\u0003ďƉ\u0003ƐƚĂƌƚ\u0003ĂŶĚ\u0003ĞŶĚ\u0003ǁŝƚŚ\u0003ƚŚĞ\u0003ďĂƐĞƐ\u0003ŝŶ\u0003LJĞůůŽǁ\u0003\nhighlight. \u0011Ϳ Top: ChIP-seq data from Zheng et al. (7) and Reese et al. (30) are displayed for this \nƌĞŐŝŽŶ͘\u0003\u0011ŽƩŽŵ͗\u0003EŽƌŵĂůŝnjĞĚ\u0003^E\u0004W\u0003ŝŶƚĞŶƐŝƟĞƐ\u0003ĨŽƌ\u0003ŐĞŶŽŵŝĐ\u0003ƉƌŽďĞƐ\u0003ĂĐƌŽƐƐ\u0003ƚŚŝƐ\u0003ƌĞŐŝŽŶ\u0003ƟůĞĚ\u0003Ăƚ\u0003every \n6 bp are displayed for this region. \n\u0003\u0003\n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted April 18, 2026. ; https://doi.org/10.64898/2026.04.16.719077doi: bioRxiv preprint \n\n32 \n \n\u0004\u0012<EKt>\u001c\u0018'\u001cD\u001cEd^ \nThe authors thank members of the Fowler Laboratory, Alarid Laboratory, and Proteovista LLC for \ntheir helpful comments on the data and manuscript. /Ŷ\u0003ƉĂƌƟĐƵůĂƌ͕\u0003ǁĞ\u0003ƚŚĂŶŬ\u0003EĂƚĂƐŚĂ\u0003^ŽůŽĚŝŶ͕\u0003\nBlue-leaf Cordes, Kelley Salem, Kaelyn Allen, and Christy Zheng ĨŽƌ\u0003ƐĐŝĞŶƟĮĐ\u0003ĚŝƐĐƵƐƐŝŽŶƐ͘\u0003 \n\u0003\n&hE\u0018/E' \nThis research was ƐƵƉƉŽƌƚĞĚ\u0003ďLJ\u0003EĂƟŽŶĂů\u0003/ŶƐƟƚƵƚĞƐ\u0003of Health͕\u0003EĂƟŽŶĂů\u0003\u0012ĂŶĐĞƌ\u0003/ŶƐƟƚƵƚĞ [grant \nnumber R01 CA260140 to ETA and AMF] with a subaward to Proteovista LLC. The funders had \nŶŽ\u0003ƌŽůĞ\u0003ŝŶ\u0003ƚŚĞ\u0003ĚĞƐŝŐŶ\u0003ŽĨ\u0003ƚŚŝƐ\u0003ƐƚƵĚLJ͖\u0003ŝŶ\u0003ƚŚĞ\u0003ĐŽůůĞĐƟŽŶ͕\u0003ĂŶĂůLJƐĞƐ͕\u0003Žƌ\u0003ŝŶƚĞƌƉƌĞƚĂƟŽŶ\u0003ŽĨ\u0003ĚĂƚĂ͖\u0003ŝŶ\u0003ƚŚĞ\u0003\nǁƌŝƟŶŐ\u0003ŽĨ\u0003ƚŚĞ\u0003ŵĂŶƵƐĐƌŝƉƚ͖\u0003Žƌ\u0003ŝŶ\u0003ƚŚĞ\u0003ĚĞĐŝƐŝŽŶ\u0003ƚŽ\u0003ƉƵďůŝƐŚ\u0003ƚŚĞ\u0003ƌĞƐƵůƚƐ͘ Funding for open access \nĐŚĂƌŐĞ͗\u0003EĂƟŽŶĂů\u0003/ŶƐƟƚƵƚĞƐ\u0003ŽĨ\u0003,ĞĂůƚŚ͘ \n \n\u0012KE&>/\u0012d\u0003K&\u0003/Ed\u001cZ\u001c^d \nThese authors have the following roles at Proteovista LLC: Mary S. Ozers is Co-Founder, Owner, \nĂŶĚ\u0003\u0012ŚŝĞĨ\u0003^ĐŝĞŶƟĮĐ\u0003KĸĐĞƌ; Christopher L. Warren is Co-Founder and Chief Biotechnology \nKĸĐĞƌ͖\u0003WĂŝŐĞ\u0003E. Messa, employee; Noah R. Nicol, employee; Keenan S. Pearson, employee. Amy \nM. &ŽǁůĞƌ\u0003ƌĞĐĞŝǀĞƐ\u0003Ŭ\u0003ĐŚĂƉƚĞƌ\u0003ƌŽLJĂůƚLJ\u0003ĨƌŽŵ\u0003\u001cůƐĞǀŝĞƌ͕\u0003/ŶĐ͘\u0003KƚŚĞƌ\u0003ĂƵƚŚŽƌƐ\u0003ĚĞĐůĂƌĞ\u0003ŶŽ\u0003ĐŽŶŇŝĐƚ\u0003ŽĨ\u0003\ninterest. \n \n\u0004hd,KZ\u0003\u0012KEdZ/\u0011hd/KE^\u0003\n\u0012ŽŶĐĞƉƚƵĂůŝnjĂƟŽŶ͗\u0003DĂƌLJ\u0003^͘\u0003KnjĞƌƐ͕\u0003\u001cůĂŝŶĞ\u0003d ͘\u0003\u0004ůĂƌŝĚ, Christopher L. Warren, Paige E. Messa, \nKeenan S. Pearson \n\u0018ĂƚĂ\u0003\u0012ƵƌĂƟŽŶ͗\u0003WĂŝŐĞ\u0003E. Messa, Christopher L. Warren, Noah R. Nicol \nFormal Analysis: Christopher L. Warren, Paige E. Messa, Noah R. Nicol, Keenan S. Pearson͕\u0003:ƵƐƟŶ\u0003\nP.  Pe t e r s  \n&ƵŶĚŝŶŐ\u0003ĂĐƋƵŝƐŝƟŽŶ͗\u0003\u001cůĂŝŶĞ\u0003d ͘\u0003\u0004ůĂƌŝĚ͕\u0003\u0004ŵLJ\u0003D͘\u0003&ŽǁůĞƌ͕\u0003DĂƌLJ\u0003^͘\u0003KnjĞƌƐ \n/ŶǀĞƐƟŐĂƟŽŶ͗ Paige E. Messa, Keenan S. Pearson, Christopher L. Warren, Noah R. Nicol \nMethodology: Paige E. Messa, Keenan S. Pearson, Christopher L. Warren, Noah R. Nicol, Mary S. \nOzers \nWƌŽũĞĐƚ\u0003\u0004ĚŵŝŶŝƐƚƌĂƟŽŶ͗ Mary S. Ozers, Elaine T. Alarid \nResources: Mary S. Ozers \n^ŽŌǁĂƌĞ͗ Noah R. Nicol, Christopher L. Warren \nSupervision: Mary S. Ozers, Elaine T. Alarid, Amy M. Fowler \nsĂůŝĚĂƟŽŶ͗ Paige E. Messa, Keenan S. Pearson, Christopher L. Warren, Noah R. Nicol \nsŝƐƵĂůŝnjĂƟŽŶ͗ Paige E. Messa, Keenan S. Pearson, Christopher L. Warren, Noah R. Nicol, Mary S. \nOzers \ntƌŝƟŶŐ\u0003– ŽƌŝŐŝŶĂů\u0003ĚƌĂŌ: Paige E. Messa, Mary S. Ozers \ntƌŝƟŶŐ\u0003– ƌĞǀŝĞǁ\u0003Θ\u0003ĞĚŝƟŶŐ͗\u0003WĂŝŐĞ\u0003E. Messa, Christopher L. Warren, Noah R. Nicol, Keenan S. \nWĞĂƌƐŽŶ͕\u0003:ƵƐƟŶ\u0003W ͘\u0003WĞƚĞƌƐ͕\u0003\u0004ŵLJ\u0003D͘\u0003&ŽǁůĞƌ͕\u0003\u001cůĂŝŶĞ\u0003d ͘\u0003\u0004ůĂƌŝĚ͕\u0003DĂƌLJ\u0003^͘\u0003KnjĞƌƐ \n \n \n\u0003\u0003\n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted April 18, 2026. ; https://doi.org/10.64898/2026.04.16.719077doi: bioRxiv preprint \n\n33 \n \nZ\u001c&\u001cZ\u001cE\u0012\u001c^\u0003\n1. \u0004ƵĚĞŶ\u0003\u0004͕\u0003\u0012ĂĚĚLJ\u0003:͕\u0003tŝůĂŶŽǁƐŬŝ\u0003d ͕\u0003dŝŶŐ\u0003^\u0011͕\u0003\u0012ƵŶŶŝŶŐŚĂŵ\u0003:D͕\u0003:ĂŶĞ\u0003^D͘\u0003^ƉĂƟĂů\u0003ĂŶĚ\u0003ƚĞŵƉŽƌĂů\u0003\nĞdžƉƌĞƐƐŝŽŶ\u0003ŽĨ\u0003ƚŚĞ\u0003'ƌĂŝŶLJŚĞĂĚ-ůŝŬĞ\u0003ƚƌĂŶƐĐƌŝƉƟŽŶ\u0003ĨĂĐƚŽƌ\u0003ĨĂŵŝůLJ\u0003ĚƵƌŝŶŐ\u0003ŵƵƌŝŶĞ\u0003ĚĞǀĞůŽƉŵĞŶƚ͘\u0003'ĞŶĞ\u0003\n\u001cdžƉƌ\u0003WĂƩĞƌŶƐ͘\u0003ϮϬϬϲ͖ϲ;ϴͿ͗ϵϲϰ-ϳϬ͘\u0003\u001cƉƵď\u0003ϮϬϬϲϬϰϮϱ͘\u0003ĚŽŝ͗\u0003ϭϬ͘ϭϬϭϲͬũ͘ŵŽĚŐĞƉ͘ϮϬϬϲ͘Ϭ3.011. PubMed \nPMID: 16831572. \n2. tŝůĂŶŽǁƐŬŝ\u0003d ͕\u0003dƵĐŬĮĞůĚ\u0003\u0004͕\u0003\u0012ĞƌƌƵƟ\u0003>͕\u0003KΖ\u0012ŽŶŶĞůů\u0003^͕\u0003^ĂŝŶƚ\u0003Z͕\u0003WĂƌĞŬŚ\u0003s͕\u0003Ğƚ\u0003Ăů͘\u0003\u0004\u0003ŚŝŐŚůLJ\u0003\nĐŽŶƐĞƌǀĞĚ\u0003ŶŽǀĞů\u0003ĨĂŵŝůLJ\u0003ŽĨ\u0003ŵĂŵŵĂůŝĂŶ\u0003ĚĞǀĞůŽƉŵĞŶƚĂů\u0003ƚƌĂŶƐĐƌŝƉƟŽŶ\u0003ĨĂĐƚŽƌƐ\u0003ƌĞůĂƚĞĚ\u0003ƚŽ\u0003\u0018ƌŽƐŽƉŚŝůĂ\u0003\ngrainyhead. Mech Dev. 2002;114(1-2):37-ϱϬ͘\u0003ĚŽŝ͗\u0003ϭϬ͘ϭϬϭϲͬƐϬϵϮϱ-4773(02)00046-1. PubMed \nPMID: 12175488. \n3. Cieply B, Riley P , Pifer PM, Widmeyer J, A ddison JB, Ivanov AV, et al. Suppression of the \nepithelial-ŵĞƐĞŶĐŚLJŵĂů\u0003ƚƌĂŶƐŝƟŽŶ\u0003ďLJ\u0003'ƌĂŝŶLJŚĞĂĚ-like-2. Cancer Res. 2012;72(9):2440-53. Epub \nϮϬϭϮϬϮϮϵ͘\u0003ĚŽŝ͗\u0003ϭϬ͘ϭϭϱϴͬϬϬϬϴ-5472.CAN-11-4038. PubMed PMID: 22379025; PubMed Central \nPMCID: PMC3342427. \n4. Cieply B, Farris J, Denvir J, Ford HL, Frisch SM. Epithelial- ŵĞƐĞŶĐŚLJŵĂů\u0003ƚƌĂŶƐŝƟŽŶ\u0003ĂŶĚ\u0003\ntumor suppression are controlled by a reciprocal feedback loop between ZEB1 and Grainyhead-\nlike-2. Cancer Res. 2013;73(20):6299-ϯϬϵ͘\u0003\u001cƉƵď\u0003ϮϬϭϯϬϴϭϯ͘\u0003ĚŽŝ͗\u0003ϭϬ͘ϭϭϱϴͬϬϬϬϴ-5472.CAN-12-\n4082. PubMed PMID: 23943797; PubMed Central PMCID: PMC3806457. \n5. Rifat Y , Parekh V, Wilanowski T, Hislop NR, Auden A, Ting SB, et al. Regional neural tube \nĐůŽƐƵƌĞ\u0003ĚĞĮŶĞĚ\u0003ďLJ\u0003ƚŚĞ\u0003'ƌĂŝŶLJ\u0003ŚĞĂĚ-ůŝŬĞ\u0003ƚƌĂŶƐĐƌŝƉƟŽŶ\u0003ĨĂĐƚŽƌƐ͘\u0003\u0018Ğǀ\u0003\u0011ŝŽů͘\u0003ϮϬϭϬ͖ϯϰϱ;ϮͿ͗Ϯϯϳ-45. \n\u001cƉƵď\u0003ϮϬϭϬϬϳϮϭ͘\u0003ĚŽŝ͗\u0003ϭϬ͘ϭϬϭϲͬũ͘LJĚďŝŽ͘ϮϬϭϬ͘Ϭϳ͘Ϭϭϳ͘\u0003WƵďDĞĚ\u0003WD/\u0018͗\u0003ϮϬϲϱϰϲϭϮ͘ \n6. tĞƌƚŚ\u0003D͕\u0003tĂůĞŶƟŶ\u0003<͕\u0003\u0004ƵĞ\u0003\u0004͕\u0003^ĐŚƂŶŚĞŝƚ\u0003:͕\u0003tƵĞďŬĞŶ\u0003\u0004͕\u0003WŽĚĞ-Shakked N, et al. The \nƚƌĂŶƐĐƌŝƉƟŽŶ\u0003ĨĂĐƚŽƌ\u0003ŐƌĂŝŶLJŚĞĂĚ-ůŝŬĞ\u0003Ϯ\u0003ƌĞŐƵůĂƚĞƐ\u0003ƚŚĞ\u0003ŵŽůĞĐƵůĂƌ\u0003ĐŽŵƉŽƐŝƟŽŶ\u0003ŽĨ\u0003ƚŚĞ\u0003ĞƉŝƚŚĞůŝĂů\u0003\nĂƉŝĐĂů\u0003ũƵŶĐƟŽŶĂů\u0003ĐŽŵƉůĞdž͘\u0003\u0018ĞǀĞůŽƉŵĞŶƚ͘\u0003ϮϬϭϬ͖ϭϯϳ;ϮϮͿ͗ϯϴϯϱ-ϰϱ͘\u0003ĚŽŝ͗\u0003ϭϬ͘ϭϮϰϮͬĚĞǀ͘Ϭϱϱϰ83. \nPubMed PMID: 20978075. \n7. Zheng C, Allen KO, Liu T, Solodin NM, Meyer MB, Salem K, et al. Elevated GRHL2 Imparts \nWůĂƐƟĐŝƚLJ\u0003ŝŶ\u0003\u001cZ-WŽƐŝƟǀĞ\u0003\u0011ƌĞĂƐƚ\u0003\u0012ĂŶĐĞƌ\u0003\u0012ĞůůƐ͘\u0003\u0012ĂŶĐĞƌƐ\u0003;\u0011ĂƐĞůͿ͘\u0003ϮϬϮϰ͖ϭϲ;ϭϲͿ͘\u0003\u001cƉƵď\u0003ϮϬϮϰϬϴϮϭ͘\u0003ĚŽŝ͗\u0003\nϭϬ͘ϯϯϵϬͬĐĂŶĐĞƌƐϭϲϭϲϮϵϬϲ͘\u0003WƵďDĞĚ\u0003WD/\u0018͗\u0003ϯϵϭϵϵϲϳϲ͖\u0003WƵďDĞĚ\u0003\u0012ĞŶƚƌĂů\u0003WD\u0012/\u0018͗\u0003WMC11353109. \n8. Werner S, Frey S, Riethdorf S, Schulze C, Alawi M, Kling L, et al. Dual roles of the \nƚƌĂŶƐĐƌŝƉƟŽŶ\u0003ĨĂĐƚŽƌ\u0003ŐƌĂŝŶLJŚĞĂĚ-like 2 (GRHL2) in breast cancer. J Biol Chem. \n2013;288(32):22993-ϯϬϬϴ͘\u0003\u001cƉƵď\u0003ϮϬϭϯϬϲϮϵ͘\u0003ĚŽŝ͗\u0003ϭϬ͘ϭϬϳϰͬũďĐ͘Dϭϭϯ͘ϰϱϲϮϵϯ͘\u0003WƵďDĞĚ\u0003WD/\u0018͗\u0003\n23814079; PubMed Central PMCID: PMC3743475. \n9. ZŝĞƚŚĚŽƌĨ\u0003^͕\u0003&ƌĞLJ\u0003^͕\u0003^ĂŶƚũĞƌ\u0003^͕\u0003^ƚŽƵƉŝĞĐ\u0003D͕\u0003KƩŽ\u0003\u0011͕\u0003ZŝĞƚŚĚŽƌĨ\u0003>͕\u0003Ğƚ\u0003Ăů͘\u0003\u0018ŝǀĞƌƐĞ\u0003ĞdžƉƌĞƐƐŝŽŶ\u0003\nƉĂƩĞƌŶƐ\u0003ŽĨ\u0003ƚŚĞ\u0003\u001cDd\u0003ƐƵƉƉƌĞƐƐŽƌ\u0003ŐƌĂŝŶLJŚĞĂĚ-ůŝŬĞ\u0003Ϯ\u0003;'Z,>ϮͿ\u0003ŝŶ\u0003ŶŽƌŵĂů\u0003ĂŶĚ\u0003ƚƵŵŽƵƌ\u0003ƟƐƐƵĞƐ͘\u0003/Ŷƚ\u0003:\u0003\nCancer. 2016;138(4):949-ϲϯ͘\u0003\u001cƉƵď\u0003ϮϬϭϱϬϵϮϱ͘\u0003ĚŽŝ͗\u0003ϭϬ͘ϭϬϬϮͬŝũĐ͘Ϯϵϴϰϭ͘\u0003WƵďDed PMID: 26355710. \n10. Reese RM, Harrison MM, Alarid ET. Grainyhead-like Protein 2: The Emerging Role in \nHormone-\u0018ĞƉĞŶĚĞŶƚ\u0003\u0012ĂŶĐĞƌƐ\u0003ĂŶĚ\u0003\u001cƉŝŐĞŶĞƟĐƐ͘\u0003\u001cŶĚŽĐƌŝŶŽůŽŐLJ͘\u0003ϮϬϭϵ͖ϭϲϬ;ϱͿ͗ϭϮϳϱ-88. doi: \nϭϬ͘ϭϮϭϬͬĞŶ͘ϮϬϭϵ-00213. PubMed PMID: 30958537. \n11. \u0004ƵĞ\u0003\u0004\n͕\u0003,ŝŶnjĞ\u0003\u0012͕\u0003tĂůĞŶƟŶ\u0003<͕\u0003ZƵīĞƌƚ\u0003:͕\u0003zƵƌƚĚĂƐ\u0003z ͕\u0003tĞƌƚŚ\u0003D͕\u0003Ğƚ\u0003Ăů͘\u0003\u0004\u0003'ƌĂŝŶLJŚĞĂĚ->ŝŬĞ\u0003ϮͬKǀŽ-\n>ŝŬĞ\u0003Ϯ\u0003WĂƚŚǁĂLJ\u0003ZĞŐƵůĂƚĞƐ\u0003ZĞŶĂů\u0003\u001cƉŝƚŚĞůŝĂů\u0003\u0011ĂƌƌŝĞƌ\u0003&ƵŶĐƟŽŶ\u0003ĂŶĚ\u0003>ƵŵĞŶ\u0003\u001cdžƉĂŶƐŝŽŶ͘\u0003:\u0003\u0004ŵ\u0003^ŽĐ\u0003\nNephrol. 2015;26(11):2704-ϭϱ͘\u0003\u001cƉƵď\u0003ϮϬϭϱϬϯϭϴ͘\u0003ĚŽŝ͗\u0003ϭϬ͘ϭϲϴϭͬ\u0004^E͘ϮϬϭϰϬϴϬϳϱϵ͘\u0003WƵbMed PMID: \n25788534; PubMed Central PMCID: PMC4625669. \n12. tĂŶŐ\u0003͕\u0003\u0012ŽďĂŶ\u0003\u0011͕\u0003tƵ\u0003,͕\u0003\u0012ŚŽƵĂƌĞĨ\u0003:͕\u0003\u0018ĂdžŝŶŐĞƌ\u0003>͕\u0003WĂƵůƐĞŶ\u0003Dd ͕\u0003Ğƚ\u0003Ăů͘\u0003GRHL2-controlled \nŐĞŶĞ\u0003ĞdžƉƌĞƐƐŝŽŶ\u0003ŶĞƚǁŽƌŬƐ\u0003ŝŶ\u0003ůƵŵŝŶĂů\u0003ďƌĞĂƐƚ\u0003ĐĂŶĐĞƌ ͘\u0003\u0012Ğůů\u0003\u0012ŽŵŵƵŶ\u0003^ŝŐŶĂů͘\u0003ϮϬϮϯ͖Ϯϭ;ϭͿ͗ϭϱ͘\u0003\u001cƉƵď\u0003\n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted April 18, 2026. ; https://doi.org/10.64898/2026.04.16.719077doi: bioRxiv preprint \n\n34 \n \nϮϬϮϯϬϭϮϯ͘\u0003ĚŽŝ͗\u0003ϭϬ͘ϭϭϴϲͬƐϭϮϵϲϰ-022-01029-5. PubMed PMID: 36691073; PubMed Central \nPMCID: PMC9869538. \n13. Chung VY , Tan TZ, Ye J, Huang RL, Lai HC, Kappei D, et al. The role of GRHL2 and \nĞƉŝŐĞŶĞƟĐ\u0003ƌĞŵŽĚĞůŝŶŐ\u0003ŝŶ\u0003ĞƉŝƚŚĞůŝĂů-ŵĞƐĞŶĐŚLJŵĂů\u0003ƉůĂƐƟĐŝƚLJ\u0003ŝŶ\u0003ŽǀĂƌŝĂŶ\u0003ĐĂŶĐĞƌ\u0003ĐĞůůƐ͘\u0003\u0012ŽŵŵƵŶ\u0003\n\u0011ŝŽů͘\u0003ϮϬϭϵ͖Ϯ͗ϮϳϮ͘\u0003\u001cƉƵď\u0003ϮϬϭϵϬϳϮϰ͘\u0003ĚŽŝ͗\u0003ϭϬ͘ϭϬϯϴͬƐϰϮϬϬϯ-019-0506-3. PubMed PMID: 31372511; \nPubMed Central PMCID: PMC6656769. \n14. tĂůĞŶƟŶ\u0003<͕\u0003,ŝŶnjĞ\u0003\u0012͕\u0003tĞƌƚŚ\u0003D͕\u0003,ĂĂƐĞ\u0003E͕\u0003sĂƌŵĂ\u0003^͕\u0003DŽƌĞůů\u0003Z͕\u0003Ğƚ\u0003Ăů͘\u0003\u0004\u0003'ƌŚůϮ-dependent \ngene network controls trophoblast branching morphogenesis. Development. 2015;142(6):1125-\nϯϲ͘\u0003ĚŽŝ͗\u0003ϭϬ͘ϭϮϰϮͬĚĞǀ͘ϭϭϯϴϮϵ͘\u0003WƵďDĞĚ\u0003WD/\u0018͗\u0003ϮϱϳϱϴϮϮϯ͖\u0003WƵďDĞĚ\u0003\u0012ĞŶƚƌĂů\u0003WD\u0012/\u0018͗\u0003WD\u0012ϲϱϭϳϴϯ3. \n15. Chi D, Singhal H, Li L, Xiao T, Liu W, Pun M, et al. Estrogen receptor signaling is \nreprogrammed during breast tumorigenesis. Proc Natl Acad Sci U S A. 2019;116(23):11437-43. \n\u001cƉƵď\u0003ϮϬϭϵϬϱϮϬ͘\u0003ĚŽŝ͗\u0003ϭϬ͘ϭϬϳϯͬƉŶĂƐ͘ϭϴϭϵϭϱϱϭϭϲ͘\u0003WƵďDĞĚ\u0003WD/\u0018͗\u0003ϯϭϭϭϬϬϬϮ͖\u0003WƵďDĞĚ\u0003\u0012ĞŶƚƌal \nPMCID: PMC6561257. \n16. 'ĂŽ\u0003y͕\u0003sŽĐŬůĞLJ\u0003\u0012D͕\u0003WĂƵůŝ\u0003& ͕\u0003EĞǁďĞƌƌLJ\u0003<D͕\u0003yƵĞ\u0003z ͕\u0003ZĂŶĚĞůů\u0003^,͕\u0003Ğƚ\u0003Ăů͘\u0003\u001cǀŝĚĞŶĐĞ\u0003ĨŽƌ\u0003ŵƵůƟƉůĞ\u0003\nroles for grainyhead-like 2 in the establishment and maintenance of human mucociliary airway \nepithelium.[corrected]. Proc Natl Acad Sci U S A. 2013;110(23):9356-61. Epub 20130520. doi: \nϭϬ͘ϭϬϳϯͬƉŶĂƐ͘ϭϯϬϳϱϴϵϭϭϬ͘\u0003WƵďDĞĚ\u0003WD/\u0018͗\u0003ϮϯϲϵϬϱϳϵ͖\u0003WƵďDĞĚ\u0003\u0012ĞŶƚƌĂů\u0003WD\u0012/\u0018͗\u0003WD\u0012ϯϲϳϳϰϱϯ͘ \n17. Pifer PM, Farris JC, Thomas AL, Stoilov P , Denvir J, Smith DM, et al. Grainyhead-like 2 \nŝŶŚŝďŝƚƐ\u0003ƚŚĞ\u0003ĐŽĂĐƟǀĂƚŽƌ\u0003ƉϯϬϬ͕\u0003ƐƵƉƉƌĞƐƐŝŶŐ\u0003ƚƵďƵůŽŐĞŶĞƐŝƐ\u0003ĂŶĚ\u0003ƚŚĞ\u0003ĞƉŝƚŚĞůŝĂů-mesenchymal \nƚƌĂŶƐŝƟŽŶ͘\u0003DŽů\u0003\u0011ŝŽů\u0003\u0012Ğůů͘\u0003ϮϬϭϲ͖Ϯϳ;ϭϱͿ͗Ϯϰϳϵ-ϵϮ͘\u0003\u001cƉƵď\u0003ϮϬϭϲϬϲϬϭ͘\u0003ĚŽŝ͗\u0003ϭϬ͘ϭϬϵϭͬŵďĐ.E16-04-0249. \nPubMed PMID: 27251061; PubMed Central PMCID: PMC4966987. \n18. Chung VY , Tan TZ, Tan M, Wong MK, Kuay KT, Yang Z, et al. GRHL2-miR-200-ZEB1 \nŵĂŝŶƚĂŝŶƐ\u0003ƚŚĞ\u0003ĞƉŝƚŚĞůŝĂů\u0003ƐƚĂƚƵƐ\u0003ŽĨ\u0003ŽǀĂƌŝĂŶ\u0003ĐĂŶĐĞƌ\u0003ƚŚƌŽƵŐŚ\u0003ƚƌĂŶƐĐƌŝƉƟŽŶĂů\u0003ƌĞŐƵůĂƟŽŶ\u0003ĂŶĚ\u0003ŚŝƐƚŽŶĞ\u0003\nŵŽĚŝĮĐĂƟŽŶ͘\u0003^Đŝ\u0003ZĞƉ͘\u0003ϮϬϭϲ͖ϲ͗ϭϵϵϰϯ͘\u0003\u001cƉƵď\u0003ϮϬϭϲϬϮϭϴ͘\u0003ĚŽŝ͗\u0003ϭϬ͘ϭϬϯϴͬƐƌĞƉϭϵϵϰϯ͘\u0003WƵďDĞĚ PMID: \n26887977; PubMed Central PMCID: PMC4757891. \n19. Jozwik KM, Chernukhin I, Serandour AA, Nagarajan S, Carroll JS. FOXA1 Directs H3K4 \nDŽŶŽŵĞƚŚLJůĂƟŽŶ\u0003Ăƚ\u0003\u001cŶŚĂŶĐĞƌƐ\u0003ǀŝĂ\u0003ZĞĐƌƵŝƚŵĞŶƚ\u0003ŽĨ\u0003ƚŚĞ\u0003DĞƚŚLJůƚƌĂŶƐĨĞƌĂƐĞ\u0003D>>ϯ͘\u0003\u0012Ğůů\u0003ZĞƉ͘\u0003\n2016;17(10):2715-Ϯϯ͘\u0003ĚŽŝ͗\u0003ϭϬ͘ϭϬϭϲͬũ͘ĐĞůƌĞƉ͘ϮϬϭϲ͘ϭϭ͘ϬϮϴ͘\u0003WƵďDĞĚ\u0003WD/\u0018͗\u0003ϮϳϵϮϲϴϳϯ͖\u0003WƵďDĞĚ\u0003\nCentral PMCID: PMC5177601. \n20. Green S, Walter P , Kumar V, Krust A, Bornert JM, Argos P , et al. Human oestrogen \nƌĞĐĞƉƚŽƌ\u0003Đ\u0018E\u0004͗\u0003ƐĞƋƵĞŶĐĞ͕\u0003ĞdžƉƌĞƐƐŝŽŶ\u0003ĂŶĚ\u0003ŚŽŵŽůŽŐLJ\u0003ƚŽ\u0003ǀ-erb-A. Nature. 1986;320(6058):134-9. \nĚŽŝ͗\u0003ϭϬ͘ϭϬϯϴͬϯϮϬϭϯϰĂϬ͘\u0003WƵďDĞĚ\u0003WD/\u0018͗\u0003ϯϳϱϰϬϯϰ͘ \n21. ^ůĞŝŐŚƚŚŽůŵ\u0003Z͕\u0003EĞŝůƐĞŶ\u0003\u0011<͕\u0003\u001cůŬŚĂƟď\u0003^͕\u0003&ůŽƌĞƐ\u0003>͕\u0003\u0018ƵŬŬŝƉĂƟ\u0003^͕\u0003ŚĂŽ\u0003Z͕\u0003Ğƚ\u0003Ăů͘\u0003WĞƌĐĞŶƚĂŐĞ\u0003ŽĨ\u0003\n,ŽƌŵŽŶĞ\u0003ZĞĐĞƉƚŽƌ\u0003WŽƐŝƟǀŝƚLJ\u0003ŝŶ\u0003\u0011ƌĞĂƐƚ\u0003\u0012ĂŶĐĞƌ\u0003WƌŽǀŝĚĞƐ\u0003WƌŽŐŶŽƐƟĐ\u0003sĂůƵĞ͗\u0003\u0004\u0003^ŝŶŐůĞ-/ŶƐƟƚƵƚĞ\u0003\nStudy. J Clin Med Res. 2021;13(1):9-ϭϵ͘\u0003\u001cƉƵď\u0003ϮϬϮϭϬϭϭϮ͘\u0003ĚŽŝ͗\u0003ϭϬ͘ϭϰϳϰϬͬũŽĐŵr4398. PubMed \nPMID: 33613796; PubMed Central PMCID: PMC7869562. \n22. Harvey JM, Clark GM, Osborne CK, Allred DC. Estrogen Receptor Status by \nImmunohistochemistry Is Superior to the Ligand-\u0011ŝŶĚŝŶŐ\u0003\u0004ƐƐĂLJ\u0003ĨŽƌ\u0003WƌĞĚŝĐƟŶŐ\u0003ZĞƐƉŽŶƐĞ\u0003ƚŽ\u0003\nAdjuvant Endocrine Therapy in Breast Cancer. J Clin Oncol. 2023;41(7):1331-8. doi: \nϭϬ͘ϭϮϬϬͬ:\u0012K͘22.02500. PubMed PMID: 36827742. \n23. Giaquinto AN, Sung H, Newman LA, Freedman RA, Smith RA, Star J, et al. Breast cancer \nƐƚĂƟƐƟĐƐ\u0003ϮϬϮϰ͘\u0003\u0012\u0004\u0003\u0012ĂŶĐĞƌ\u0003:\u0003\u0012ůŝŶ͘\u0003ϮϬϮϰ͖ϳϰ;ϲͿ͗ϰϳϳ-ϵϱ͘\u0003\u001cƉƵď\u0003ϮϬϮϰϭϬϬϭ͘\u0003ĚŽŝ͗\u0003ϭϬ͘ϯϯϮϮͬĐĂĂĐ͘Ϯϭϴϲϯ͘\u0003\nPubMed PMID: 39352042. \n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted April 18, 2026. ; https://doi.org/10.64898/2026.04.16.719077doi: bioRxiv preprint \n\n35 \n \n24. \u0004ůŝ\u0003^͕\u0003DĞƚnjŐĞƌ\u0003\u0018͕\u0003\u0011ŽƌŶĞƌƚ\u0003:D͕\u0003\u0012ŚĂŵďŽŶ\u0003W ͘\u0003DŽĚƵůĂƟŽŶ\u0003ŽĨ\u0003ƚƌĂŶƐĐƌŝƉƟŽŶĂů\u0003ĂĐƟǀĂƟŽŶ\u0003ďLJ\u0003\nligand-ĚĞƉĞŶĚĞŶƚ\u0003ƉŚŽƐƉŚŽƌLJůĂƟŽŶ\u0003ŽĨ\u0003ƚŚĞ\u0003ŚƵŵĂŶ\u0003ŽĞƐƚƌŽŐĞŶ\u0003ƌĞĐĞƉƚŽƌ\u0003\u0004ͬ\u0011\u0003ƌĞŐŝŽŶ͘\u0003\u001cD\u0011K\u0003:͘\u0003\n1993;12(3):1153-ϲϬ͘\u0003ĚŽŝ͗\u0003ϭϬ͘ϭϬϬϮͬũ͘ϭϰϲϬ-ϮϬϳϱ͘ϭϵϵϯ͘ƚďϬϱϳϱϲ͘dž͘\u0003WƵďDĞĚ\u0003WD/\u0018͗\u0003ϴϰϱϴϯϮϴ͖\u0003\nPubMed Central PMCID: PMC413317. \n25. \u0011ƵŶŽŶĞ\u0003'͕\u0003\u0011ƌŝĂŶĚ\u0003W\u0004͕\u0003DŝŬƐŝĐĞŬ\u0003Z:͕\u0003WŝĐĂƌĚ\u0003\u0018͘\u0003\u0004ĐƟǀĂƟŽŶ\u0003ŽĨ\u0003ƚŚĞ\u0003ƵŶůŝŐĂŶĚĞĚ\u0003ĞƐƚƌŽŐĞŶ\u0003\nƌĞĐĞƉƚŽƌ\u0003ďLJ\u0003\u001c'&\u0003ŝŶǀŽůǀĞƐ\u0003ƚŚĞ\u0003D\u0004W\u0003ŬŝŶĂƐĞ\u0003ƉĂƚŚǁĂLJ\u0003ĂŶĚ\u0003ĚŝƌĞĐƚ\u0003ƉŚŽƐƉŚŽƌLJůĂƟŽŶ͘\u0003\u001cD\u0011K\u0003:͘\u0003\n1996;15(9):2174-83. PubMed PMID: 8641283; PubMed Central PMCID: PMC450141. \n26. \u0012ŚĞŶŐ\u0003:͕\u0003ŚĂŶŐ\u0003\u0012͕\u0003^ŚĂƉŝƌŽ\u0003\u0018:͘\u0003\u0004\u0003ĨƵŶĐƟŽŶĂů\u0003ƐĞƌŝŶĞ\u0003ϭϭϴ\u0003ƉŚŽƐƉŚŽƌLJůĂƟŽŶ\u0003ƐŝƚĞ\u0003ŝŶ\u0003ĞƐƚƌŽŐĞŶ\u0003\nreceptor-alpha is required for down-ƌĞŐƵůĂƟŽŶ\u0003ŽĨ\u0003ŐĞŶĞ\u0003ĞdžƉƌĞƐƐŝŽŶ\u0003ďLJ\u0003ϭϳďĞƚĂ-estradiol and 4-\nŚLJĚƌŽdžLJƚĂŵŽdžŝĨĞŶ͘\u0003\u001cŶĚŽĐƌŝŶŽůŽŐLJ͘\u0003ϮϬϬϳ͖ϭϰϴ;ϭϬͿ͗ϰϲϯϰ-41. Epub 20070705. doi: \n1Ϭ͘ϭϮϭϬͬĞŶ͘ϮϬϬϳ-0148. PubMed PMID: 17615152. \n27. sĂůůĞLJ\u0003\u0012\u0012͕\u0003DĠƟǀŝĞƌ\u0003Z͕\u0003^ŽůŽĚŝŶ\u0003ED͕\u0003&ŽǁůĞƌ\u0003\u0004D͕\u0003DĂƐŚĞŬ\u0003Dd ͕\u0003,ŝůů\u0003>͕\u0003Ğƚ\u0003Ăů͘\u0003\u0018ŝīĞƌĞŶƟĂů\u0003\nƌĞŐƵůĂƟŽŶ\u0003ŽĨ\u0003ĞƐƚƌŽŐĞŶ-ŝŶĚƵĐŝďůĞ\u0003ƉƌŽƚĞŽůLJƐŝƐ\u0003ĂŶĚ\u0003ƚƌĂŶƐĐƌŝƉƟŽŶ\u0003ďLJ\u0003ƚŚĞ\u0003ĞƐƚƌŽŐĞŶ\u0003ƌĞĐĞƉƚŽƌ\u0003ĂůƉŚĂ\u0003E\u0003\nterminus. Mol Cell Biol. 2005;25(13):5417-Ϯϴ͘\u0003ĚŽŝ͗\u0003ϭϬ͘ϭϭϮϴͬD\u0012\u0011͘Ϯϱ͘ϭϯ͘ϱϰϭϳ-5428.2005. \nPubMed PMID: 15964799; PubMed Central PMCID: PMC1156995. \n28. Dutertre M, Smith CL. Ligand- ŝŶĚĞƉĞŶĚĞŶƚ\u0003ŝŶƚĞƌĂĐƟŽŶƐ\u0003ŽĨ\u0003ƉϭϲϬͬƐƚĞƌŽŝĚ\u0003ƌĞĐĞƉƚŽƌ\u0003\nĐŽĂĐƟǀĂƚŽƌƐ\u0003ĂŶĚ\u0003\u0012Z\u001c\u0011-binding protein (CBP) with estrogen receptor-ĂůƉŚĂ͗\u0003ƌĞŐƵůĂƟŽŶ\u0003ďLJ\u0003\nƉŚŽƐƉŚŽƌLJůĂƟŽŶ\u0003ƐŝƚĞƐ\u0003ŝŶ\u0003ƚŚĞ\u0003\u0004ͬ\u0011\u0003ƌĞŐŝŽŶ\u0003ĚĞƉĞŶĚƐ\u0003ŽŶ\u0003ŽƚŚĞƌ\u0003ƌĞĐĞƉƚŽƌ\u0003ĚŽŵĂŝŶƐ͘\u0003DŽů\u0003\u001cŶĚŽĐƌŝŶŽl. \n2003;17(7):1296-ϯϭϰ͘\u0003\u001cƉƵď\u0003ϮϬϬϯϬϰϮϰ͘\u0003ĚŽŝ͗\u0003ϭϬ͘ϭϮϭϬͬŵĞ͘ϮϬϬϭ-0316. PubMed PMID: 12714702. \n29. Helzer KT, Szatkowski Ozers M, Meyer MB, Benkusky NA, Solodin N, Reese RM, et al. The \nPhosphorylated Estrogen Receptor ɲ\u0003;\u001cZͿ\u0003\u0012ŝƐƚƌŽŵĞ\u0003/ĚĞŶƟĮĞƐ\u0003Ă\u0003^ƵďƐĞƚ\u0003ŽĨ\u0003\u0004ĐƟǀĞ\u0003\u001cŶŚĂŶĐĞƌƐ\u0003\nEnriched for Direct ER-\u0018E\u0004\u0003\u0011ŝŶĚŝŶŐ\u0003ĂŶĚ\u0003ƚŚĞ\u0003dƌĂŶƐĐƌŝƉƟŽŶ\u0003&ĂĐƚŽƌ\u0003'Z,>Ϯ͘\u0003DŽů\u0003\u0012Ğůů\u0003Biol. \nϮϬϭϵ͖ϯϵ;ϯͿ͘\u0003\u001cƉƵď\u0003ϮϬϭϵͬϬϭͬϭϲ͘\u0003ĚŽŝ͗\u0003ϭϬ͘ϭϭϮϴͬD\u0012\u0011͘ϬϬϰϭϳ-18. PubMed PMID: 30455249; PubMed \nCentral PMCID: PMC6336141. \n30. Reese RM, Helzer KT, Allen KO, Zheng C, Solodin N, Alarid ET. GRHL2 Enhances \nWŚŽƐƉŚŽƌLJůĂƚĞĚ\u0003\u001cƐƚƌŽŐĞŶ\u0003ZĞĐĞƉƚŽƌ\u0003;\u001cZͿ\u0003\u0012ŚƌŽŵĂƟŶ\u0003\u0011ŝŶĚŝŶŐ\u0003ĂŶĚ\u0003ZĞŐƵůĂƚĞƐ\u0003\u001cZ-Mediated \ndƌĂŶƐĐƌŝƉƟŽŶĂů\u0003\u0004ĐƟǀĂƟŽŶ\u0003ĂŶĚ\u0003ZĞƉƌĞƐƐŝŽŶ͘\u0003DŽů\u0003\u0012Ğůů\u0003\u0011ŝŽů͘\u0003ϮϬϮϮ͖ϰϮ;ϭϬͿ͗ĞϬϬϭϵϭϮϮ͘\u0003\u001cƉƵď\u0003\nϮϬϮϮϬϴϮϵ͘\u0003ĚŽŝ͗\u0003ϭϬ͘ϭϭϮϴͬŵĐď͘ϬϬϭϵϭ-22. PubMed PMID: 36036613; PubMed Central PMCID: \nPMC9584124. \n31. Hurtado A, Holmes KA, Ross-Innes CS, Schmidt D, Carroll JS. FOXA1 is a key determinant \nŽĨ\u0003ĞƐƚƌŽŐĞŶ\u0003ƌĞĐĞƉƚŽƌ\u0003ĨƵŶĐƟŽŶ\u0003ĂŶĚ\u0003ĞŶĚŽĐƌŝŶĞ\u0003ƌĞƐƉŽŶƐĞ͘\u0003EĂƚ\u0003'ĞŶĞƚ͘\u0003ϮϬϭϭ͖ϰϯ;ϭͿ͗Ϯϳ-33. doi: \nϭϬ͘ϭϬϯϴͬŶŐ͘ϳϯϬ͘\u0003WƵďDĞĚ\u0003WD/\u0018͗\u0003ϮϭϭϱϭϭϮϵ͖\u0003WƵďDĞĚ\u0003\u0012ĞŶƚƌĂů\u0003WD\u0012/\u0018͗\u0003ϯϬϮϰϱϯϳ͘ \n32. ^ĞĂĐŚƌŝƐƚ\u0003\u0018\u0018͕\u0003\u0004ŶƐƟŶĞ\u0003>:͕\u0003<Ğƌŝ\u0003Z\u0004͘\u0003&Ky\u0004ϭ͗\u0003\u0004\u0003WŝŽŶĞĞƌ\u0003ŽĨ\u0003EƵĐůĞĂƌ\u0003ZĞĐĞƉƚŽƌ\u0003\u0004ĐƟŽŶ\u0003ŝŶ\u0003\u0011ƌĞĂƐƚ\u0003\n\u0012ĂŶĐĞƌ ͘\u0003\u0012ĂŶĐĞƌƐ\u0003;\u0011ĂƐĞůͿ͘\u0003ϮϬϮϭ͖ϭϯ;ϮϬͿ͘\u0003\u001cƉƵď\u0003ϮϬϮϭϭϬϭϳ͘\u0003ĚŽŝ͗\u0003ϭϬ͘ϯϯϵϬͬĐĂŶĐĞƌƐϭϯϮϬϱϮϬϱ͘\u0003WƵďDĞĚ\u0003\nPMID: 34680352; PubMed Central PMCID: PMC8533709. \n33. KnjĞƌƐ\u0003D^͕\u0003tĂƌƌĞŶ\u0003\u0012>͕\u0003\u0004ŶƐĂƌŝ\u0003\u0004͘\u0003\u0018ĞƚĞƌŵŝŶŝŶŐ\u0003\u0018E\u0004\u0003ƐĞƋƵĞŶĐĞ\u0003ƐƉĞĐŝĮĐŝƚLJ\u0003ŽĨ\u0003ŶĂƚƵƌĂů\u0003ĂŶĚ\u0003\nĂƌƟĮĐŝĂů\u0003ƚƌĂŶƐĐƌŝƉƟŽŶ\u0003ĨĂĐƚŽƌƐ\u0003ďLJ\u0003ĐŽŐŶĂƚĞ\u0003ƐŝƚĞ\u0003ŝĚĞŶƟĮĞƌ\u0003ĂŶĂůLJƐŝƐ͘\u0003DĞƚŚŽĚƐ\u0003DŽů\u0003\u0011ŝŽů͘\u0003\n2009;544:637-ϱϯ͘\u0003ĚŽŝ͗\u0003ϭϬ͘ϭϬϬϳͬϵϳϴ-1-59745-483-ϰͺϰϭ͘\u0003WƵďDĞĚ\u0003WD/\u0018͗\u0003ϭϵϰϴϴϳϮϵ͘ \n34. ZĂũďŚĂŶĚĂƌŝ\u0003W\n͕\u0003KnjĞƌƐ\u0003D^͕\u0003^ŽůŽĚŝŶ\u0003ED͕\u0003tĂƌƌĞŶ\u0003\u0012>͕\u0003\u0004ůĂƌŝĚ\u0003\u001cd ͘\u0003WĞƉƟĚLJůƉƌŽůLJů\u0003/ƐŽŵĞƌĂƐĞ\u0003\nWŝŶϭ\u0003\u0018ŝƌĞĐƚůLJ\u0003\u001cŶŚĂŶĐĞƐ\u0003ƚŚĞ\u0003\u0018E\u0004\u0003\u0011ŝŶĚŝŶŐ\u0003&ƵŶĐƟŽŶƐ\u0003ŽĨ\u0003\u001cƐƚƌŽŐĞŶ\u0003ZĞĐĞƉƚŽƌ\u0003ɲ͘\u0003J Biol Chem. \n2015;290(22):13749-ϲϮ͘\u0003ĚŽŝ͗\u0003ϭϬ͘ϭϬϳϰͬũďĐ͘Dϭϭϰ͘ϲϮϭϲϵϴ͘\u0003WƵďDĞĚ\u0003WD/\u0018͗\u0003ϮϱϴϲϲϮϬϵ͖\u0003WƵďDĞĚ\u0003\nCentral PMCID: PMC4447953. \n35. Holding AN, Giorgi FM, Donnelly A, Culle n AE, Nagarajan S, Selth LA, et al. VULCAN \nintegrates ChIP-ƐĞƋ\u0003ǁŝƚŚ\u0003ƉĂƟĞŶƚ-derived co-ĞdžƉƌĞƐƐŝŽŶ\u0003ŶĞƚǁŽƌŬƐ\u0003ƚŽ\u0003ŝĚĞŶƟĨLJ\u0003'Z,>Ϯ\u0003ĂƐ\u0003Ă\u0003ŬĞLJ\u0003ĐŽ-\n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted April 18, 2026. ; https://doi.org/10.64898/2026.04.16.719077doi: bioRxiv preprint \n\n36 \n \nregulator of ERa at enhancers in breast cancer. Genome Biol. 2019;20(1):91. Epub 20190513. \nĚŽŝ͗\u0003ϭϬ͘ϭϭϴϲͬƐϭϯϬϱϵ-019-1698-z. PubMed PMID: 31084623; PubMed Central PMCID: \nPMC6515683. \n36. Berger MF, Philippakis AA, Qureshi AM, He  FS, Estep PW, 3rd, Bulyk ML. Compact, \nƵŶŝǀĞƌƐĂů\u0003\u0018E\u0004\u0003ŵŝĐƌŽĂƌƌĂLJƐ\u0003ƚŽ\u0003ĐŽŵƉƌĞŚĞŶƐŝǀĞůLJ\u0003ĚĞƚĞƌŵŝŶĞ\u0003ƚƌĂŶƐĐƌŝƉƟŽŶ-factor binding site \nƐƉĞĐŝĮĐŝƟĞƐ͘\u0003EĂƚ\u0003\u0011ŝŽƚĞĐŚŶŽů͘\u0003ϮϬϬϲ͖Ϯϰ;ϭϭͿ͗ϭϰϮϵ-ϯϱ͘\u0003ĚŽŝ͗\u0003ϭϬ͘ϭϬϯϴͬŶďƚϭϮϰϲ͘\u0003WƵďDĞĚ\u0003WD/\u0018͗ \n16998473. \n37. Berger MF, Bulyk ML. Universal protein-binding microarrays for the comprehensive \nĐŚĂƌĂĐƚĞƌŝnjĂƟŽŶ\u0003ŽĨ\u0003ƚŚĞ\u0003\u0018E\u0004-ďŝŶĚŝŶŐ\u0003ƐƉĞĐŝĮĐŝƟĞƐ\u0003ŽĨ\u0003ƚƌĂŶƐĐƌŝƉƟŽŶ\u0003ĨĂĐƚŽƌƐ͘\u0003EĂƚ\u0003WƌŽƚŽĐ͘\u0003\n2009;4(3):393-ϰϭϭ͘\u0003ĚŽŝ͗\u0003ϭϬ͘ϭϬϯϴͬŶƉƌŽƚ͘ϮϬϬϴ͘ϭϵϱ͘\u0003WƵďDĞĚ\u0003WD/\u0018͗\u0003ϭϵϮϲϱϳϵϵ͖\u0003WƵďDĞĚ\u0003\u0012ĞŶƚƌĂů\u0003\nPMCID: PMC2908410. \n38. WĞƚĞƌƐ\u0003:W ͕\u0003DĂŚĞƌ\u0003>:͘\u0003\u0018E\u0004\u0003ĐƵƌǀĂƚƵƌĞ\u0003ĂŶĚ\u0003ŇĞdžŝďŝůŝƚLJ\u0003ŝŶ\u0003ǀŝƚƌŽ\u0003ĂŶĚ\u0003ŝŶ\u0003ǀŝǀŽ͘\u0003Y\u0003ZĞǀ\u0003\u0011ŝŽƉŚLJƐ͘\u0003\n2010;43(1):23-ϲϯ͘\u0003\u001cƉƵď\u0003ϮϬϭϬϬϱϭϴ͘\u0003ĚŽŝ͗\u0003ϭϬ͘ϭϬϭϳͬ^ϬϬϯϯϱϴϯϱϭϬϬϬϬϬϳϳ͘\u0003WƵďDĞĚ\u0003WD/\u0018͗\u0003\n20478077; PubMed Central PMCID: PMC4190679. \n39. Becker NA, Greiner AM, Peters JP , Maher LJ. Bacterial promoter repression by DNA \nlooping without protein-ƉƌŽƚĞŝŶ\u0003ďŝŶĚŝŶŐ\u0003ĐŽŵƉĞƟƟŽŶ͘\u0003EƵĐůĞŝĐ\u0003\u0004ĐŝĚƐ\u0003ZĞƐ͘\u0003ϮϬϭϰ͖ϰϮ;ϵͿ͗ϱϰϵϱ-504. \n\u001cƉƵď\u0003ϮϬϭϰϬϯϬϱ͘\u0003ĚŽŝ͗\u0003ϭϬ͘ϭϬϵϯͬŶĂƌͬŐŬƵϭϴϬ͘\u0003WƵďDĞĚ\u0003WD/\u0018͗\u0003ϮϰϱϵϴϮϱϲ͖\u0003WƵďDĞĚ\u0003\u0012ĞŶƚƌĂů\u0003WD\u0012/\u0018͗\u0003\nPMC4027209. \n40. ŚĂŶŐ\u0003z ͕\u0003DĐ\u001cǁĞŶ\u0003\u0004\u001c͕\u0003\u0012ƌŽƚŚĞƌƐ\u0003\u0018D͕\u0003>ĞǀĞŶĞ\u0003^\u0018͘\u0003^ƚĂƟƐƟĐĂů-mechanical theory of DNA \nlooping. Biophys J. 2006;90(6):1903-ϭϮ͘\u0003\u001cƉƵď\u0003ϮϬϬϱϭϮϭϲ͘\u0003ĚŽŝ͗\u0003ϭϬ͘ϭϱϮϵͬďŝŽƉŚLJƐũ͘ϭϬϱ͘ϬϳϬϰϵϬ͘\u0003\nPubMed PMID: 16361335; PubMed Central PMCID: PMC1386771. \n41. hƐŚŝĚĂ\u0003\u0012͕\u0003\u0004ŝďĂ\u0003,͘\u0003,ĞůŝĐĂů\u0003ƉŚĂƐĞ\u0003ĚĞƉĞŶĚĞŶƚ\u0003ĂĐƟŽŶ\u0003ŽĨ\u0003\u0012ZW͗\u0003ĞīĞĐƚ\u0003ŽĨ\u0003ƚŚĞ\u0003ĚŝƐƚĂŶĐĞ\u0003ďĞƚǁĞĞŶ\u0003\nthe CRP site and the -ϯϱ\u0003ƌĞŐŝŽŶ\u0003ŽŶ\u0003ƉƌŽŵŽƚĞƌ\u0003ĂĐƟǀŝƚLJ͘\u0003EƵĐůĞŝĐ\u0003\u0004ĐŝĚƐ\u0003ZĞƐ͘\u0003ϭϵϵϬ͖ϭϴ;ϮϭͿ͗ϲϯϮϱ-30. \nĚŽŝ͗\u0003ϭϬ͘ϭϬϵϯͬŶĂƌͬϭϴ͘Ϯϭ͘ϲϯϮϱ͘\u0003WƵďDĞĚ\u0003WD/\u0018͗\u0003ϮϭϳϯϴϮϲ͖\u0003WƵďDĞĚ\u0003\u0012ĞŶƚƌĂů\u0003WD\u0012/\u0018͗\u0003PMC332499. \n42. \u0012ĂƌůƐŽŶ\u0003\u0012\u0018͕\u0003tĂƌƌĞŶ\u0003\u0012>͕\u0003,ĂƵƐĐŚŝůĚ\u0003<\u001c͕\u0003KnjĞƌƐ\u0003D^͕\u0003YĂĚŝƌ\u0003E͕\u0003\u0011ŚŝŵƐĂƌŝĂ\u0003\u0018͕\u0003Ğƚ\u0003Ăů͘\u0003^ƉĞĐŝĮĐŝƚLJ\u0003\nůĂŶĚƐĐĂƉĞƐ\u0003ŽĨ\u0003\u0018E\u0004\u0003ďŝŶĚŝŶŐ\u0003ŵŽůĞĐƵůĞƐ\u0003ĞůƵĐŝĚĂƚĞ\u0003ďŝŽůŽŐŝĐĂů\u0003ĨƵŶĐƟŽŶ͘\u0003WƌŽĐ\u0003EĂƚů\u0003\u0004ĐĂĚ\u0003^Đŝ\u0003h\u0003^\u0003\u0004͘\u0003\n2010;107(10):4544-ϵ͘\u0003ĚŽŝ͗\u0003ϭϬ͘ϭϬϳϯͬƉŶĂƐ͘ϬϵϭϰϬϮϯϭϬϳ͘\u0003WƵďDĞĚ\u0003WD/\u0018͗\u0003ϮϬϭϳϲϵϲϰ͖\u0003PubMed \nCentral PMCID: PMC2842033. \n43. DŝŶŐ\u0003Y͕\u0003ZŽƐŬĞ\u0003z ͕\u0003^ĐŚƵĞƚnj\u0003\u0004͕\u0003tĂůĞŶƟŶ\u0003<͕\u0003/ďƌĂŝŵŝ\u0003/͕\u0003^ĐŚŵŝĚƚ-KƩ\u0003<D͕\u0003Ğƚ\u0003Ăů͘\u0003^ƚƌƵĐƚƵƌĂů\u0003ďĂƐŝƐ\u0003\nŽĨ\u0003ŐĞŶĞ\u0003ƌĞŐƵůĂƟŽŶ\u0003ďLJ\u0003ƚŚĞ\u0003'ƌĂŝŶLJŚĞĂĚͬ\u0012WϮ\u0003ƚƌĂŶƐĐƌŝƉƟŽŶ\u0003ĨĂĐƚŽƌ\u0003ĨĂŵŝůLJ͘\u0003EƵĐůĞŝĐ\u0003\u0004ĐŝĚƐ\u0003ZĞƐ͘\u0003\n2018;46(4):2082-ϵϱ͘\u0003ĚŽŝ͗\u0003ϭϬ͘ϭϬϵϯͬŶĂƌͬŐŬdžϭϮϵϵ͘\u0003WƵďDĞĚ\u0003WD/\u0018͗\u0003ϮϵϯϬϵϲϰϮ͖\u0003WƵďDĞĚ Central \nPMCID: PMC5829564. \n44. Fan C, Wang Q, van der Zon G, Ren J, Agaser C, Slieker RC, et al. OVOL1 inhibits breast \nĐĂŶĐĞƌ\u0003ĐĞůů\u0003ŝŶǀĂƐŝŽŶ\u0003ďLJ\u0003ĞŶŚĂŶĐŝŶŐ\u0003ƚŚĞ\u0003ĚĞŐƌĂĚĂƟŽŶ\u0003ŽĨ\u0003d'&-ɴ\u0003type I receptor. Signal Transduct \ndĂƌŐĞƚ\u0003dŚĞƌ ͘\u0003ϮϬϮϮ͖ϳ;ϭͿ͗ϭϮϲ͘\u0003\u001cƉƵď\u0003ϮϬϮϮϬϰϮϵ͘\u0003ĚŽŝ͗\u0003ϭϬ͘ϭϬϯϴͬƐϰϭϯϵϮ-022-00944-w. PubMed PMID: \n35484112; PubMed Central PMCID: PMC9050647. \n45. Drago-'ĂƌĐŝĂ\u0003\u0018͕\u0003'ŝƌŝ\u0003^͕\u0003\u0012ŚĂƩĞƌũĞĞ\u0003Z͕\u0003^ŝŵŽŶŝ-Nieves A, Abedrabbo M, Genna A, et al. Re-\nĞƉŝƚŚĞůŝĂůŝnjĂƟŽŶ\u0003ŽĨ\u0003ĐĂŶĐĞƌ\u0003ĐĞůůƐ\u0003ŝŶĐƌĞĂƐĞƐ\u0003ĂƵƚŽƉŚĂŐLJ\u0003ĂŶĚ\u0003\u0018E\u0004\u0003ĚĂŵĂŐĞ͗\u0003/ŵƉůŝĐĂƟŽŶƐ\u0003ĨŽƌ\u0003ďƌĞĂƐƚ\u0003\nc\nancer dormancy and relapse. Sci Signal. 2025;18(883):eado3473. Epub 20250422. doi: \nϭϬ͘ϭϭϮϲͬƐĐŝƐŝŐŶĂů͘ĂĚŽϯϰϳϯ͘\u0003WƵďDĞĚ\u0003WD/\u0018͗\u0003ϰϬϮϲϭϵϱϱ͖\u0003WƵďDĞĚ\u0003\u0012ĞŶƚƌĂů\u0003WD\u0012/\u0018͗\u0003WD\u0012ϭϮϭϯϱϴϯϯ͘ \n46. \u0011ĂŝůĞLJ\u0003d>͕\u0003\u0011ŽĚĞŶ\u0003D͕\u0003\u0011ƵƐŬĞ\u0003&\u0004͕\u0003&ƌŝƚŚ\u0003D͕\u0003'ƌĂŶƚ\u0003\u0012\u001c͕\u0003\u0012ůĞŵĞŶƟ\u0003>͕\u0003Ğƚ\u0003Ăů͘\u0003D\u001cD\u001c\u0003^h/d\u001c͗\u0003ƚŽŽůƐ\u0003ĨŽƌ\u0003\nŵŽƟĨ\u0003ĚŝƐĐŽǀĞƌLJ\u0003ĂŶĚ\u0003ƐĞĂƌĐŚŝŶŐ͘\u0003EƵĐůĞŝĐ\u0003\u0004ĐŝĚƐ\u0003ZĞƐ͘\u0003ϮϬϬϵ͖ϯϳ;tĞď\u0003^ĞƌǀĞƌ\u0003ŝƐƐƵĞͿ͗tϮϬϮ-8. Epub \nϮϬϬϵͬϬϱͬϮϬ͘\u0003ĚŽŝ͗\u0003ϭϬ͘ϭϬϵϯͬŶĂƌͬŐŬƉϯϯϱ͘\u0003WƵďDĞĚ\u0003WD/\u0018͗\u0003ϭϵϰϱϴϭϱϴ͖\u0003WƵďDĞĚ\u0003\u0012ĞŶƚƌĂů\u0003WDCID: \nPMC2703892. \n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted April 18, 2026. ; https://doi.org/10.64898/2026.04.16.719077doi: bioRxiv preprint \n\n37 \n \n47. >ŝ\u0003^͕\u0003KůƐŽŶ\u0003t<͕\u0003>Ƶ\u0003y:͘\u0003tĞď\u0003ϯ\u0018E\u0004\u0003Ϯ͘Ϭ\u0003ĨŽƌ\u0003ƚŚĞ\u0003ĂŶĂůLJƐŝƐ͕\u0003ǀŝƐƵĂůŝnjĂƟŽŶ͕\u0003ĂŶĚ\u0003ŵŽĚĞůŝŶŐ\u0003ŽĨ\u0003ϯ\u0018\u0003\nnucleic acid structures. Nucleic Acids Res. 2019;47(W1):W26-tϯϰ͘\u0003ĚŽŝ͗\u0003ϭϬ͘ϭϬϵϯͬŶĂƌͬŐŬnjϯϵϰ͘\u0003\nPubMed PMID: 31114927; PubMed Central PMCID: PMC6602438. \n48. Glont SE, Chernukhin I, Carroll JS. Comprehensive Genomic Analysis Reveals that the \nWŝŽŶĞĞƌŝŶŐ\u0003&ƵŶĐƟŽŶ\u0003ŽĨ\u0003&Ky\u0004ϭ\u0003/Ɛ\u0003/ŶĚĞƉĞŶĚĞŶƚ\u0003ŽĨ\u0003,ŽƌŵŽŶĂů\u0003^ŝŐŶĂůŝŶŐ͘\u0003\u0012Ğůů\u0003ZĞƉ͘\u0003\n2019;26(10):2558-ϲϱ͘Ğϯ͘\u0003ĚŽŝ͗\u0003ϭϬ͘ϭϬϭϲͬũ͘ĐĞůƌĞƉ͘ϮϬϭϵ͘ϬϮ͘Ϭϯϲ͘\u0003WƵďDĞĚ\u0003WD/\u0018͗\u0003ϯϬϴϰϬϴϴϭ͖\u0003WƵďDĞĚ\u0003\nCentral PMCID: PMC6408623. \n49. 'ůŽŶƚ\u0003^\u001c͕\u0003WĂƉĂĐŚƌŝƐƚŽƵ\u0003\u001c<͕\u0003^ĂǁůĞ\u0003\u0004͕\u0003,ŽůŵĞƐ\u0003<\u0004͕\u0003\u0012ĂƌƌŽůů\u0003:^͕\u0003^ŝĞƌƐďĂĞŬ\u0003Z͘\u0003/ĚĞŶƟĮĐĂƟŽŶ\u0003ŽĨ\u0003\nChIP-ƐĞƋ\u0003ĂŶĚ\u0003Z/D\u001c\u0003ŐƌĂĚĞ\u0003ĂŶƟďŽĚŝĞƐ\u0003ĨŽƌ\u0003\u001cƐƚƌŽŐĞŶ\u0003ZĞĐĞƉƚŽƌ\u0003ĂůƉŚĂ͘\u0003W>Ž^\u0003KŶĞ͘\u0003\nϮϬϭϵ͖ϭϰ;ϰͿ͗ĞϬϮϭϱϯϰϬ͘\u0003\u001cƉƵď\u0003ϮϬϭϵϬϰϭϬ͘\u0003ĚŽŝ͗\u0003ϭϬ͘ϭϯϳϭͬũŽƵƌŶĂů͘ƉŽŶĞ͘ϬϮϭϱϯϰϬ͘\u0003WƵďDĞĚ\u0003WD/\u0018͗\u0003\n30970003; PubMed Central PMCID: PMC6457525. \n50. \u0004ĂƌƚƐ\u0003Dd ͕\u0003EŽƌĚŝŶ\u0003\u0004͕\u0003\u0012ĂŶƚƶ\u0003\u0012͕\u0003ǀĂŶ\u0003\u0011ŽdžƚĞů\u0003\u0004>͕\u0003ǀĂŶ\u0003\u0004ŵĞƌŽŶŐĞŶ\u0003Z͘\u0003DĞĐŚĂŶŝƐƟĐ\u0003ĚŝƐƐĞĐƟŽŶ\u0003ŽĨ\u0003\n'Z,>Ϯ\u0003ĂŶĚ\u0003WZ\u0003ƚƌĂŶƐĐƌŝƉƟŽŶĂů\u0003ĐŽ-ƌĞŐƵůĂƟŽŶ\u0003ŝŶ\u0003ďƌĞĂƐƚ\u0003ĐĞůůƐ͘\u0003W>Ž^\u0003'ĞŶĞƚ͘\u0003ϮϬϮϲ͖ϮϮ;ϯͿ͗ĞϭϬϭϮϬϴϴ͘\u0003\n\u001cƉƵď\u0003ϮϬϮϲϬϯϭϳ͘\u0003ĚŽŝ͗\u0003ϭϬ͘ϭϯϳϭͬũŽƵƌŶĂů͘ƉŐĞŶ͘ϭϬϭϮϬϴϴ͘\u0003WƵďDĞĚ\u0003WD/\u0018͗\u0003ϰϭϴϰϯϲϮϮ͖\u0003WƵbMed \nCentral PMCID: PMC13008249. \n51. Mohammed H, Taylor C, Brown GD, Papach ristou EK, Carroll JS, D'Santos CS. Rapid \nŝŵŵƵŶŽƉƌĞĐŝƉŝƚĂƟŽŶ\u0003ŵĂƐƐ\u0003ƐƉĞĐƚƌŽŵĞƚƌLJ\u0003ŽĨ\u0003ĞŶĚŽŐĞŶŽƵƐ\u0003ƉƌŽƚĞŝŶƐ\u0003;Z/D\u001cͿ\u0003ĨŽƌ\u0003ĂŶĂůLJƐŝƐ\u0003ŽĨ\u0003\nĐŚƌŽŵĂƟŶ\u0003ĐŽŵƉůĞdžĞƐ͘\u0003EĂƚ\u0003WƌŽƚŽĐ͘\u0003ϮϬϭϲ͖ϭϭ;ϮͿ͗ϯϭϲ-Ϯϲ͘\u0003\u001cƉƵď\u0003ϮϬϭϲͬϬϭͬϮϭ͘\u0003ĚŽŝ͗\u0003\nϭϬ͘ϭϬϯϴͬŶƉƌŽƚ͘ϮϬϭ6.020. PubMed PMID: 26797456. \n \n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted April 18, 2026. ; https://doi.org/10.64898/2026.04.16.719077doi: bioRxiv preprint \n\nGraphical Abstract\n .CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted April 18, 2026. ; https://doi.org/10.64898/2026.04.16.719077doi: bioRxiv preprint \n\nGRHL2 625 aa\nTransactivation domain\nDNA binding domain (DBD)\nConserved DBD segment\nDimerization domain\n1           1 3 3      2 4 7                    4 8 4  ~ 4 9 4         6 2 5\n416  424\nFigure 1 .CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted April 18, 2026. ; https://doi.org/10.64898/2026.04.16.719077doi: bioRxiv preprint \n\nA\nB\n9.163\n11.826\n3.003\n0.879\nSNAP \nFluorescence \nIntensity\nSNAP Array Sequence\nChromosome 11\n……………………………………………….....\nChIP-Seq\n0.532\n4.577\nOVOL1\n5’-CCCACCTATTTGTTACCTGTCGAACCGGTTTCCATTCCGCTGCGGGTGCATCGCAAGCCT-3’\n3’-GGGTGGATAAACAATGGACAGCTTGGCCAAAGGTAAGGCGACGCCCACGTAGCGTTCGGA-5’\n5’-TGACAACCCACCTATTTGTTACCTGTCGAACCGGTTTCCATTCCGCTGCATCGCAAGCCT-3’\n3’-ACTGTTGGGTGGATAAACAATGGACAGCTTGGCCAAAGGTAAGGCGACGTAGCGTTCGGA-5’\n5’-TAACGGTGACAACCCACCTATTTGTTACCTGTCGAACCGGTTTCCATTCATCGCAAGCCT-3’\n3’-ATTGCCACTGTTGGGTGGATAAACAATGGACAGCTTGGCCAAAGGTAAGTAGCGTTCGGA-5’\n5’-GGGAAATAACGGTGACAACCCACCTATTTGTTACCTGTCGAACCGGTTCATCGCAAGCCT-3’\n3’-CCCTTTATTGCCACTGTTGGGTGGATAAACAATGGACAGCTTGGCCAAGTAGCGTTCGGA-5’\nP3: chr11_65786865-65786912\nP2: chr11_65786859-65786906\nP4: chr11_65786871-65786918\nP5: chr11_65786877-65786924\n5’-TGCGTTGGGAAATAACGGTGACAACCCACCTATTTGTTACCTGTCGAACATCGCAAGCCT-3’\n3’-ACGCAACCCTTTATTGCCACTGTTGGGTGGATAAACAATGGACAGCTTGTAGCGTTCGGA-5’\nP6: chr11_65786883-65786930\nP1: chr11_65786853-65786900\nAdd purified GRHL2 \nto SNAP array\nDetect on \nSNAP array\nGenomic sequence\nPrimer\nLinker\nSlide\n5’-TATTTGTTACCTGTCGAACCGGTTTCCATTCCGCTGCGGGTGAGGTGGCATCGCAAGCCT-3’\n3’-ATAAACAATGGACAGCTTGGCCAAAGGTAAGGCGACGCCCACTCCACCGTAGCGTTCGGA-5’\nD\nArray features each containing 105 –1 06\ncopies of dsDNA probes968,014 total DNA features\nC\nBound features\ndsDNA probes\nFigure 2\n .CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted April 18, 2026. ; https://doi.org/10.64898/2026.04.16.719077doi: bioRxiv preprint \n\n0\n5\n10\n15\n20\n0\n5\n10\n15\n20\n_GA AACCGGTT C_GAG\n1\n2\n4\n5\n7\n8\nB\nC\nE\n Position 1    A 1,984 mutations (4.4%) \n Position 2    A 1,917 mutations (4.2%) \n Position 3   C 861 mutations (1.9%) \n Position 4   C 17,930 mutations (39.5%) \n Position 5    G 17,930 mutations (39.5%) \n Position 6    G 861 mutations (1.9%) \n Position 7    T 1,917 mutations (4.2%) \n Position 8    T 1,984 mutations (4.4%)\nDip in 0 mm ring:\nConsensus at either end of 48mer\n(Edge or next to primer)\nCA repeats\n(low complexity)\nTA repeats followed by GA repeats\n(low complexity)\nAACNGGTT\n(middle area of probe)\nAACNNGTT\n(best = AACWWGTT)\nD\nA\n 1-300\n 301-600\nPosition\nBits\n0\n1\n2\n601-900 901-1200\nPosition\nBits\n0\n1\n2\nPosition\nBits\n0\n1\n2\nPosition\nBits\n0\n1\n-1 1 2 3 4\n2\n5678 + 1\n3\n6__\nLocation\nPreferred\nBase\nEnrichment\nRatio P-value\n2 bp upstream\n1 bp upstream\n1 bp downstream\n3 bp downstream\n4 bp downstream\n5 bp downstream\nG\nA\nC\nG\nA\nG\n1.55x\n1.71x\n1.59x\n1.28x\n1.62x\n1.44x\n<0.001***\n<0.001***\n0.003**\n0.042*\n<0.001***\n<0.001***\nPosition\nBits\n0\n1\n-5 -4 -3 -2 -1\n2\n1 2 3 4 5 6 7 8 +1 +2 +3 +4 +5\nFigure 3\n- 1123456 78-2\n- 112345678-2 - 112345678-2\n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted April 18, 2026. ; https://doi.org/10.64898/2026.04.16.719077doi: bioRxiv preprint \n\nC1-AACCGGTT-C2\nMismatches\nNormalized Intensity\nConsensus 1 2 3 4\n2\n4\n6\n0\nBC\nMismatches (Left_Right)\nNormalized Intensity\n1_1 2_2 3_3 4_4 1_0 2_0 0_3 0_40_1 0_23_0 4_0\n10\n0\nD C1-AACCGGTT-A-AACCGGTT-C2\nMotif\nNormalized Intensity\nGACCGGTT\n0\n2\n4\n6\nTACCGGTT\nAACGGGTT\nCACCGGTT\nAACCGTTT\nAACCGGAT\nAACCTGTT\nAGCCGGTT\nACCCGGTT\nAACTGGTT\nAATCGGTT\nAACCGGTT\nE\nNormalized Intensity\n0\n2\n4\n6\nAACCGGTT\nAACCGGTC\nAACCGGTA\nTACCGGTA\nGGCCGGCA\nGGCCGGCC\nGACCGGCC\nCCCCCGTA\nCAGCGGCC\nAAACTTTG\nATGCCGTA\nAAGCAGAG\nATCCCCTG\nATGCCGTG\nF\n1-1200\nPosition\nBits\n0\n1\n2345\n2\n67 8 12345678\nA\nFigure 4\nMotif 1 Motif 2\n0_0\n30\n20\n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted April 18, 2026. ; https://doi.org/10.64898/2026.04.16.719077doi: bioRxiv preprint \n\nFigure 5\n .CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted April 18, 2026. ; https://doi.org/10.64898/2026.04.16.719077doi: bioRxiv preprint \n\nA\nchrX:120,603,600-120,604,450\nNormalized Tag Density\nEndogenous\nGRHL2\nGRHL2 \nOver-\nexpression\n200 bp\nMCTS1\n25\n150\n0\n10\n20\n30 SNAP\nArray\nRelative GRHL2\nBinding Intensity\nDirect Binding Event Indirect Binding Event\nchr22:47,631,400-47,632,250\n25\n150\n200 bp\nEPIC1\n0\n10\n20\n30 SNAP\nArray\nRelative GRHL2\nBinding Intensity\nB\n772,732 Genomic SNAP Probes\n6,151 Genomic Regions\n4,566 GRHL2 Bound\nDirect binding\n1,585 Unbound\nIndirect binding\nC\n4000\n3000\n2000\n1000\n0\n5102 3 4\n1,585\n3,814\n668\n72 11 1\nNumber of Distinct GRHL2 Binding Events\nNumber of Genomic Regions\nNumber of Distinct GRHL2 Binding Events \nIn SNAP Array Genomic Regions\nD\nDirect Binding Event Indirect Binding Event\nSNAP Array:\nSignal: YES\nSignal: NOSignal: YES\nChIP-seq:\nSNAP Array:\nChIP-seq:\nSignal: YES\nNormalized Tag Density\nFigure 6 .CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted April 18, 2026. ; https://doi.org/10.64898/2026.04.16.719077doi: bioRxiv preprint \n\nA\nB\nATTGCTTGGCTGAGGTTGCCAGGTTTCTCCACCGTAAAGTTACTTTTT\nTTTTTTTTAAAGGTGTGCAAATTAAATGAGTGGTTTCTCATAGTCAAAC\nAACCGGTTTCTCATAGCCAAACAACCGGTTTCTCATAGCCAAACAAC\nCGGTTTCTCATAGTCAAACAACCGGTTTCTCATAGTCAAACAACCGG\nTTTCTCATAGTCAAACAACCGGTTTCTCATAGCCAAACAACCGGTTTC\nTCATAGTCAAACAACCGGTTTCTCATAGTCAAACAACCGGTTTCTCAT\nAGTCAAACAACCGGTTTCTAATAGTCAAACAACCGGTTTCTCATAGTC\nAAACAACCGGTTTCTCATAGCCAAACAACCGGTTTCTCATAGCCAAA\nCAACCGGTTTCTCATAGTCAAACAACTGGTTTCTCATAGTCAAACAAC\nCGGTTTCTCATAGCCAAACAACCGGTTTCTCATAGCCAAACAACCGG\nTTTCTCATAGTCAAACAACCGGTTTCTCATAGTCAAACAACCGGTTTC\nTCATAGTCAAACAACCGGTTTCTCATAGTCAAACAAACGGTTTCTCAT\nAGTCAAACAACCGGTTTCTCATAGTCAAACAACCGGTTTCTCATAGTC\nAAACAAACGGTTTCTCATAGTCAAACAACCGGTTTCTAATAGTCAAAC\nACTTTCCTCTGGGCTGGCTTCATTCTCATGCTCTCAGTGGCAAAGAG\nAGCTCCTAGTAGAAAACTCAATGAGAAAAGTGAAGAACA\nchr7:155,123,400-155,125,400\nNormalized Tag Density\nEndogenous\nGRHL2\nGRHL2 \n24-hour \nInduction\n200 bp\n25\n150\n10\n20\n30 SNAP\nArray\nRelative GRHL2\nBinding Intensity\nFigure 7 .CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted April 18, 2026. ; https://doi.org/10.64898/2026.04.16.719077doi: bioRxiv preprint","source_license":"CC-BY-4.0","license_restricted":false}