Prediction Of YY1 Loop Anchor Based On Multi-Omics Features

Chromatin is a complex structure that plays a critical role in regulating gene expression.In recent years,research has focused on the three-dimensional organization of chromatin and its impact on gene expression.The transcription factor YY1 is known to promote interactions between enhancers and promoters,similar to the chromatin interactions mediated by CTCF.YY1 plays an important role in gene expression and cell fate determination.However,it is not clear which YY1 binding sites can form loop anchors.This study integrates multiple omics datasets,including genomics,transcriptomics,and epigenomics data,using the Light GBM model to predict YY1 loop anchors.Different types of data can provide distinct information,and integrating these diverse data sets can offer a more comprehensive understanding of YY1’s mechanism of action,leading to more accurate predictions of YY1 loop anchors.To evaluate the performance of the model,we used the area under the precision-recall curve(AUPRC)as a metric,which is suitable for imbalanced data and large differences in positive and negative samples.The results showed that our Light GBM model had good predictive performance(AUPRC ≥ 0.93).We also tested the robustness of the model by dividing the dataset into training and independent test sets in a 4:1 ratio.The results showed that our model performed well on both the training and independent test sets and could effectively predict YY1 loop anchors.Our research results indicate that histone modifications,conservation analysis,co-binding of transcription factors,and specific DNA sequences are all associated with the formation of loop structures in YY1-mediated chromatin interactions.The Ch IP-Seq signal intensity of YY1 binding sites that form loop anchors is significantly higher than that of sites without anchor formation.In addition,we ranked the feature importance and found that the co-binding of transcription factors CTCF,SMC3,RAD21,histone modifications,and sequence context are the main factors that influence the formation of YY1 loop anchor and sequence context are the main factors that influence the formation of YY1 loop anchor points.This finding will contribute to a better understanding of the mechanism of YY1-mediated chromatin interactions in future research.Innovations:(1)We propose a novel method that integrates multiple omics data and utilizes machine learning algorithms to predict YY1 loop anchors.This approach is of great significance for understanding the mechanisms of chromatin three-dimensional organization and gene expression regulation.(2)We found that histone modifications,conservation analysis,co-binding of transcription factors,and specific DNA sequences are associated with YY1 loop anchor formation.Co-binding of transcription factors CTCF,SMC3,RAD21,histone modifications,and sequence context were identified as the main factors that affect YY1 loop anchor formation.These findings provide new directions and insights for future in-depth studies on the mechanisms of YY1-mediated chromatin interactions.