RESUMEN
The advent of Neural-network Quantum States (NQS) has significantly advanced wave function ansatz research, sparking a resurgence in orbital space variational Monte Carlo (VMC) exploration. This work introduces three algorithmic enhancements to reduce computational demands of VMC optimization using NQS: an adaptive learning rate algorithm, constrained optimization, and block optimization. We evaluate the refined algorithm on complex multireference bond stretches of H2O and N2 within the cc-pVDZ basis set and calculate the ground-state energy of the strongly correlated chromium dimer (Cr2) in the Ahlrichs SV basis set. Our results achieve superior accuracy compared to coupled cluster theory at a relatively modest CPU cost. This work demonstrates how to enhance optimization efficiency and robustness using these strategies, opening a new path to optimize large-scale restricted Boltzmann machine-based NQS more effectively and marking a substantial advancement in NQS's practical quantum chemistry applications.
RESUMEN
The androgen receptor signaling inhibitor (ARSI) enzalutamide (Enz) has shown critical efficacy in the treatment of advanced prostate cancer (PCa). However, the development of drug resistance is a significant factor contributing to mortality in PCa patients. We aimed to explore the key mechanisms of Enz-resistance. Through analysis of GEO databases, we identified SLC4A4 as a novel driver in Enz resistance. Long-term Enz treatment leads to the up-regulation of SLC4A4, which in turn mediates P53 lactylation via the NF-κB/STAT3/SLC4A4 axis, ultimately leading to the development of Enz resistance and progression of PCa. SLC4A4 knockdown overcomes Enz resistance both in vitro and in vivo. Hence, our results suggest that targeting SLC4A4 could be a promising therapeutic strategy for Enz resistance. STATEMENT OF SIGNIFICANCE: SLC4A4 is a novel driver of enzalutamide resistance.