Review of molecular biological studies on acute lymphoblastic leukemia treated by modified shengmaiyin.

The objective was to explore the pharmacological mechanism of modified shengmaiyin (MSMY) in the treatment of acute lymphoblastic leukemia (ALL) by network pharmacology analysis. The effective components and predicted targets of MSMY were collected from TCMSP and Swiss target prediction databases, and the related targets of ALL were screened by GeneCards and DisGeNET. The core targets and related signaling pathways of MSMY active ingredients for the treatment of ALL were predicted by protein-protein interaction network (PPI), gene ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) functional enrichment analysis. We identified 172 potential targets for the active components of MSMY, 538 disease targets associated with ALL, and 59 common gene targets. PPI network showed that 27 targets such as triptolide, RAC-alpha serine/threonine-protein kinase (AKT1), vascular endothelial growth factor A and Caspase-3 (CASP3) were the core targets. KEGG enrichment analysis related signaling pathways included cancer pathway, phosphatidylinositol 3 kinase, PI-3K/protein kinase B (PI3K-Akt) signaling pathway, apoptosis and mitogen-activated protein kinase (MAPK) signaling pathway and IL-17 signaling pathway. The effective active components and potential therapeutic targets of MSMY in the treatment of ALL were initially identified by comprehensive network pharmacology, which provides a theoretical basis for further study of the material basis and molecular mechanism of MSMY in the treatment of ALL.

von Hippel-Lindau-targeting microRNA-143-3p attenuates mitochondrial abnormality via AMPK/PGC-1α axis in Parkinson's disease.

Parkinson's disease (PD) is the second most common neurodegenerative disease characterized by selective loss of dopaminergic neurons. We previously found that inhibition of von Hippel-Lindau (VHL) can alleviate dopaminergic neuron degeneration in PD models via regulation of mitochondrial homeostasis, however, the disease-related alterations of VHL and the regulatory mechanisms of VHL level in PD need to be further investigated. In this study, we found that the levels of VHL were markedly increased in multiple cell models of PD and identified microRNA-143-3p (miR-143-3p) as a promising candidate for regulating VHL expression involved in PD. miR-143-3p directly bound to the 3'untranslated region of human VHL mRNA and inhibited its translation, and exerted neuroprotective effects by improving cell viability, apoptosis and tyrosine hydroxylase abnormality. Furthermore, we demonstrated that miR-143-3p exerted neuroprotection by attenuating mitochondrial abnormality via AMP-activated protein kinase (AMPK)/peroxisome proliferator-activated receptor γ coactivator-1α (PGC-1α) axis, and AMPK inhibitor abolished the beneficial effects of miR-143-3p on the cell model of PD. Therefore, we identify the dysregulated VHL and miR-143-3p in PD, and propose the therapeutic potential of miR-143-3p to alleviate PD by improving mitochondrial homeostasis via AMPK/PGC-1α axis.

Major-Minor Long Short-Term Memory for Word-Level Language Model.

Language model (LM) plays an important role in natural language processing (NLP) systems, such as machine translation, speech recognition, learning token embeddings, natural language generation, and text classification. Recently, the multilayer long short-term memory (LSTM) models have been demonstrated to achieve promising performance on word-level language modeling. For each LSTM layer, larger hidden size usually means more diverse semantic features, which enables the LM to perform better. However, we have observed that when a certain LSTM layer reaches a sufficiently large scale, the promotion of overall effect will slow down, as its hidden size increases. In this article, we analyze that an important factor leading to this phenomenon is the high correlation between the newly extended hidden states and the original hidden states, which hinders diverse feature expression of the LSTM. As a result, when the scale is large enough, simply lengthening the LSTM hidden states will cost tremendous extra parameters but has little effect. We propose a simple yet effective improvement on each LSTM layer consisting of a large-scale Major LSTM and a small-scale Minor LSTM to break the high correlation between the two parts of hidden states, which we call Major-Minor LSTMs (MMLSTMs). In experiments, we demonstrate the LM with MMLSTMs surpasses the existing state-of-the-art model on Penn Treebank (PTB) and WikiText-2 (WT2) data sets and outperforms the baseline by 3.3 points in perplexity on WikiText-103 data set without increasing model parameter counts.