Abstract Clonal hematopoiesis (CH), such as clonal hematopoiesis of indeterminant potential (CHIP), is diagnosed based on somatic genomic alterations in the absence of hematologic malignancy. At present, CHIP is diagnosed using peripheral blood where putative driver point mutations and small insertions/deletions whose variant allele frequency is greater or equal to two percent. Generally, the prevalence of CH increases as an individual ages and conveys a risk for progression to a malignancy. Previously, we developed a method using fluctuating CpG (fCpG) sites to serve as a fluctuating methylation clock to uncover stem cell dynamics in glandular tissues and orthogonally validated our method using publicly available datasets of human blood from normal cohorts and malignant cohorts. Here we expand on this work by presenting 38 new patients with distinct VAF groups from 1-2% VAF up to greater than 10% VAF for putative drivers with corresponding DNA methylation profiles using the Illumina EPIC array platform. We identify fCpG from our normal and CHIP cohorts to train and validate a machine learning method that allows us to diagnose CHIP without DNA sequencing. This method allows for the identification of patients who may have CH driven by structural variants such as copy number variants. We use this method to examine a cohort of 656 normal patients without evidence of CHIP based on mutation showing that we identify the presence of CHIP roughly a third of these patients. Clonal hematopoiesis is driven by the underlying hematopoietic stem cells of an unknown quantity, with estimates for stim cell numbers differing by orders of magnitude.Finally, using a mechanistic model of clonal hematopoeisis and fCpGs we examine the temporal ability to diagnose CHIP given single and multiple expanding clones. We illustrate that multiple driver mutations within the same clone leads to an increased detection ability with an increased expansion rate (i.e. fitness) and in cases where drivers are found in separate, competing clones. Citation Format: Ryan O. Schenck, Niels A. Jakobsen, Virginia Turati, Darryl Shibata, Paresh Vyas, Simon Leedham, Alexander R. Anderson. Mutation agnostic diagnosis of clonal hematopoiesis of indeterminate potential using fluctuating methylation clocks [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2022; 2022 Apr 8-13. Philadelphia (PA): AACR; Cancer Res 2022;82(12_Suppl):Abstract nr 634.