Biology subjects

Ni, N.

Publications and source records attributed to Ni, N..

6 recordsLinked to original sources

bioRxiv2025

Nanopore direct RNA sequencing (DRS) of MS2 bacteriophages in E.coli throughout its life cycle reveals a complex transcriptional activity to control and maintain its growth

The RNA bacteriophage MS2 is an RNA phage that infects the bacterium E. coli and is one of the most studied and prototypical model phages in molecular biology and microbiology. Previous research revealed complex translational control and fine-tuning for MS2 replication. However, the dynamics of its transcriptional activity and replication during the life cycles within the bacteria remain elusive. Here, we employed Nanopore Direct RNA sequencing (DRS) to investigate the transcriptome and epitranscriptome landscape of the MS2 in infected E. coli throughout multiple life cycles. We discovered that MS2 phages sustain a high level of transcriptional activity required for replication. We found large amounts of subgenomic small transcripts from RNA degradation, Nanopore DRS bias, and transcripts containing the coat-encoding region, required for virion assembly. We found hybrid reads due to the error-prone activity of the MS2 replicase complex by the template-switching mechanism. We evidenced that RNA modification is conserved throughout the entire life cycle in full-length transcripts without the acquisition of new modifications, whereas small transcripts did acquire newly modified sites. The conserved sequence and secondary structure (U-rich hairpin) of {Psi} installation sites were the most amenable to RNA modification, and one site aligned well with host RluA-mediated installation. Overall, our investigation reveals a more complex transcriptional dynamics of MS2 phages than anticipated within E. coli to maintain its growth and replicate under host pressure.

microbiology↗

bioRxiv2023

A versatile framework for drug-target interaction prediction by considering domain specific features

Predicting drug-target interaction (DTI) is a critical and rate-limiting step in drug discovery. Traditional wet-lab experiments are reliable but expensive and time-consuming. Recently, deep learning has revealed itself as a new and promising tool for accelerating the DTI prediction process because its powerful performance. Due to the vast chemical space, the DTI prediction models are typically expected to discover drugs or targets that are absent from the training set. However, generalizing prediction performance to novel drug-target pairs that belong to different distributions is a challenge for deep learning methods. In this work, we propose an Ensemble of models that capture both Domain-generIc and domain-Specific features (E-DIS) to learn diversity domain features and adapt to out-of-distribution (OOD) data. We employed Mixture-of-Experts (MOE) as a domain-specific feature extractor for the raw data to prevent the loss of any crucial features by the encoder during the learning process. Multiple experts are trained on different domains to capture and align domain-specific information from various distributions without accessing any data from unseen domains. We evaluate our approach using four benchmark datasets under both in-domain and cross-domain settings and compare it with advanced approaches for solving OOD generalization problems. The results demonstrate that E-DIS effectively improves the robustness and generalizability of DTI prediction models by incorporating diversity domain features.

bioinformatics↗

bioRxiv2023

Assisting and Accelerating NMR Assignment with Restrainted Structure Prediction

NMR experiments can detect in situ structures and dynamic interactions, but the NMR assignment process requires expertise and is time-consuming, thereby limiting its applicability. Deep learning algorithms have been employed to aid in experimental data analysis. In this work, we developed a RASP model which can enhance structure prediction with restraints. Based on the Evoformer and structure module architecture of AlphaFold, this model can predict structure based on sequence and a flexible number of input restraints. Moreover, it can evaluate the consistency between the predicted structure and the imposed restraints. Based on this model, we constructed an iterative NMR NOESY peak assignment pipeline named FAAST, to accelerate assignment process of NOESY restraints and obtaining high quality structure ensemble. The RASP model and FAAST pipeline not only allow for the leveraging of experimental restraints to improve model prediction, but can also facilitate and expedite experimental data analysis with their integrated capabilities.

bioinformatics↗

bioRxiv2022

PanGu Drug Model: Learn a Molecule Like a Human

Recent achievements in large-scale pre-trained models like GPT-3 and PanGu- showed amazing performances in many downstream tasks, which makes AI friendlier toward industrial users. Deep learning has been recognized as the most promising technology for pharmaceuticals, a powerful molecule pre-trained model could save researchers tons of time. In chemistry classes, the students learn two molecule representations, the molecular formula and the structure formula, and learn to translate them from one way to the other. Inspired by this, we developed a novel deep learning architecture using a graph-to-sequence asymmetric conditional variational autoencoders, called PanGu Drug Model, which can appropriately characterize the molecule from both representations and improve the performance of downstream drug discovery tasks. After pretrained with 1.7 billion small molecules, our proposed model achieved the state-of-the-art results in 20 drug discovery tasks, such as molecule property prediction (predict ADMET properties, compound-target interactions, drug-drug interactions and chemical reaction productivity), molecule generation and molecule optimization. A new drug screening library of 100 million drug-like small molecules with 99.68% novelty was generated by PanGu molecule generator, which could efficiently produce novel compounds with similar physiochemical properties to given distribution, this library could be used to supplement existing compound databases. In addition, PanGu molecule optimizer could optimize the chemical structures of starting molecule with improved molecular property of interest. An automatic multi-objective optimization web application implemented by PanGu Drug Model is available at http://www.pangu-drug.com/.

biochemistry↗

bioRxiv2019

Early urinary protein changes in the tumor-forming process of the NuTu-19 lung metastasis

The early detection of cancer is essential for effective intervention. Urine, which lacks homeostatic control, has been used to reflect early systemic changes in subcutaneous cancer models. However, urine has not been used to predict whether tumors will be formed in animal models. In this study, a cancer model was established by the tail-vein injection of 2 million NuTu-19 tumor cells. Approximately half of the rats formed lung metastatic carcinoma tumors. Urine samples were randomly selected from 4 tumor-forming rats and 4 non-tumor-forming rats on day 0/12/27/39/52 and analyzed by label-free proteomic quantitative analysis. Compared to day 0, a total of 180 and 118 urinary proteins in tumor-forming and non-tumor-forming rats, respectively, showed significant changes. Functional enrichment analysis of the differential proteins in tumor-forming rats revealed similar events during cancer metastasis cascades and tumor progression, such as the migration of tumor cell lines, coagulation system, TGF-{beta} signaling, the STAT3 pathway and the adhesion of myeloid cells and alveolar macrophages. The differential proteins in non-tumor-forming rats were associated with agranulocyte/granulocyte adhesion and diapedesis, glutathione biosynthesis, IL-12 signaling and vitamin metabolism. After validation by parallel reaction monitoring (PRM) targeted proteomics quantitative analysis, a total of 22 urinary proteins showed significant changes in the early phase of the lung tumor formation process, 5 of which have been associated with the mechanisms of lung cancer, while 3 have been suggested as lung cancer biomarkers. Another 14 urinary proteins showed significant changes in the early phase in non-tumor-forming rats. Our results indicate that urine proteins could differentiate early tumor-forming and non-tumor-forming status.

cancer biology↗

bioRxiv2018

Early candidate biomarkers in urine of Walker-256 lung metastasis rat model.

Cancer metastasis accounts for the majority of deaths by cancer. Detection of cancer metastasis at its early stage is important for the management and prediction of cancer progression. Urine, which is not regulated by homeostatic mechanisms, reflects systemic changes in the whole body and can potentially be used for the early detection of cancer metastasis. In this study, a lung metastasis of a Walker-256 rat model was established by tail-vein injection of Walker-256 cells. Urine samples were collected at days 2, 4, 6 and 9 after injection, and the urinary proteomes were profiled using liquid chromatography coupled with tandem mass spectrometry (LC-MS/MS). The urinary protein patterns changed significantly with the development of Walker-256 lung metastasis. On the fourth day, lung metastasis nodules appeared. On the sixth day, clinical symptoms started. On days 2, 4, 6 and 9, 11, 25, 34 and 44 differential proteins were identified in 7 lung metastatic rats by LC-MS/MS. Seventeen of these 62 differential proteins were identified on the second day, and 18 of them were identified on the fourth day. The differential urinary proteins changed significantly two days before lung metastasis nodules appeared. Differential urinary proteins differed in Walker-256 lung metastasis rat models and Walker-256 subcutaneous rat models. A total of 9 differential proteins (NHRF1, CLIC1, EZRI, AMPN, ACY1A, HSP7C, BTD, NID2, and CFAD) were identified in 7 lung metastatic rats at one or more common time points, and these 9 differential proteins were not identified in the subcutaneous rat model. Seven of these 9 differential proteins were associated with both breast cancer and lung cancer, eight of the nine were identified on the second day, and 8 of the nine can be identified on the fourth day; these early changes in urine were also identified with differential abundances at late stages of lung metastasis. Our results indicate that (1) the urine proteome changed significantly, even on the second day after tail-vein injection of Walker-256 cells and that (2) the urinary differential proteins were different in Walker-256 lung metastatic tumors and Walker-256 subcutaneous tumors. Our results provide the potential to detect early breast cancer lung metastasis, monitor its progression and differentiate it from the same cancer cells grown at other locations.

biochemistry↗