Discordance between chromatin accessibility and transcriptional activity during the human primed-to-naïve pluripotency transition process

YY1 binding is a gene-intrinsic barrier to Xist-mediated gene silencing

X chromosome inactivation (XCI) in mammals is mediated by Xist RNA which functions in cis to silence genes on a single X chromosome in XX female cells, thereby equalising levels of X-linked gene expression relative to XY males. XCI progresses over a period of several days, with some X-linked genes silencing faster than others. The chromosomal location of a gene is an important determinant of silencing rate, but uncharacterised gene-intrinsic features also mediate resistance or susceptibility to silencing. In this study, we examine mouse embryonic stem cell lines with an inducible Xist allele (iXist-ChrX mESCs) and integrate allele-specific data of gene silencing and decreasing inactive X (Xi) chromatin accessibility over time courses of Xist induction with cellular differentiation. Our analysis reveals that motifs bound by the transcription factor YY1 are associated with persistently accessible regulatory elements, including many promoters and enhancers of slow-silencing genes. We further show that YY1 is evicted relatively slowly from target sites on Xi, and that silencing of X-linked genes is increased upon YY1 degradation. Together our results suggest that YY1 acts as a barrier to Xist-mediated silencing until the late stages of the XCI process.

SynCluster: Reaction Type Clustering and Recommendation Framework for Synthesis Planning

AI-assisted synthesis planning has emerged as a valuable tool in accelerating synthetic chemistry for the discovery of new drugs and materials. The template-free approach, which showcases superior generalization capabilities, is seen as the mainstream direction in this field. However, it remains unclear whether such an end-to-end approach can achieve problem-solving performance on par with experienced chemists without fully revealing insights into the chemical mechanisms involved. Moreover, there is a lack of unified and chemically inspired frameworks for improving multitask reaction predictions in this area. In this study, we have addressed these challenges by investigating the impact of fine-grained reaction-type labels on multiple downstream tasks and propose a novel framework named SynCluster. This framework incorporates unsupervised clustering cues into the baseline models and identifies plausible chemical subspaces which is compatible with multitask extensions and can serve as model-independent indicators to effectively enhance the performance of multiple downstream tasks. In retrosynthesis prediction, SynCluster achieves significant improvements of 4.1 and 11.0% in top-1 and top-10 prediction accuracy, respectively, compared to the baseline Molecular Transformer, and achieves a notable enhancement of 13.9% in top-10 accuracy when combined with Retroformer. By incorporating simplified molecular-input line-entry system augmentation, our framework achieves higher top-10 accuracy compared to state-of-the-art sequence-based retrosynthesis models and improves over the baseline on the diversity and validity of reactants. SynCluster also achieves 94.9% top-10 accuracy in forward synthesis prediction and 51.5% top-10 Maxfrag accuracy in reagent prediction. Overall, SynCluster provides a fresh perspective with chemical interpretability and reinforcement of domain knowledge in the synthesis design. It offers a promising solution for improving the accuracy and efficiency of AI-assisted synthesis planning and bridges the gap between template-free approaches and the problem-solving abilities of experienced chemists.