RESUMO
Increasing the binding affinity of an antibody to its target antigen is a crucial task in antibody therapeutics development. This paper presents a pretrainable geometric graph neural network, GearBind, and explores its potential in in silico affinity maturation. Leveraging multi-relational graph construction, multi-level geometric message passing and contrastive pretraining on mass-scale, unlabeled protein structural data, GearBind outperforms previous state-of-the-art approaches on SKEMPI and an independent test set. A powerful ensemble model based on GearBind is then derived and used to successfully enhance the binding of two antibodies with distinct formats and target antigens. ELISA EC50 values of the designed antibody mutants are decreased by up to 17 fold, and KD values by up to 6.1 fold. These promising results underscore the utility of geometric deep learning and effective pretraining in macromolecule interaction modeling tasks.
Assuntos
Afinidade de Anticorpos , Redes Neurais de Computação , Humanos , Anticorpos/imunologia , Anticorpos/química , Simulação por Computador , Aprendizado Profundo , Antígenos/imunologia , Ligação Proteica , Ensaio de Imunoadsorção Enzimática , Modelos MolecularesRESUMO
The advent of single-cell RNA sequencing (scRNA-seq) technologies has revolutionized transcriptomic studies. However, large-scale integrative analysis of scRNA-seq data remains a challenge largely due to unwanted batch effects and the limited transferabilty, interpretability, and scalability of the existing computational methods. We present single-cell Embedded Topic Model (scETM). Our key contribution is the utilization of a transferable neural-network-based encoder while having an interpretable linear decoder via a matrix tri-factorization. In particular, scETM simultaneously learns an encoder network to infer cell type mixture and a set of highly interpretable gene embeddings, topic embeddings, and batch-effect linear intercepts from multiple scRNA-seq datasets. scETM is scalable to over 106 cells and confers remarkable cross-tissue and cross-species zero-shot transfer-learning performance. Using gene set enrichment analysis, we find that scETM-learned topics are enriched in biologically meaningful and disease-related pathways. Lastly, scETM enables the incorporation of known gene sets into the gene embeddings, thereby directly learning the associations between pathways and topics via the topic embeddings.
Assuntos
Bases de Dados Genéticas , Modelos Genéticos , Análise de Sequência de RNA/estatística & dados numéricos , Análise de Célula Única/métodos , Doença de Alzheimer/genética , Doença de Alzheimer/patologia , Animais , Transtorno Depressivo Maior/genética , Transtorno Depressivo Maior/patologia , Perfilação da Expressão Gênica/métodos , Perfilação da Expressão Gênica/estatística & dados numéricos , Genes Mitocondriais , Humanos , Camundongos , Redes Neurais de Computação , RNA Citoplasmático Pequeno , Retina/citologia , Retina/fisiologia , Análise de Sequência de RNA/métodosRESUMO
To solve the issues of the poor dispersion performance of the inorganic flame retardant filler in epoxy resin (EP) matrix, the three-dimensional (3D) hybrid carbon nanotubes-copper aluminum (sodium dodecyl sulfate)-layered double hydroxide (CNTs-OLDH) is designed and synthesized by co-precipitation. The results indicate that the CNTs-OLDH hybrid with 3D-structure is successfully fabricated and well dispersed in EP matrix. The thermostability of EP/CNTs-OLDH nanocomposites is raised and the residue is obviously increased. When the amount of CNTs-OLDH is only 4 wt%, limit oxygen index value of EP/CNTs-OLDH nanocomposites reaches 28.5. Compared with pure EP, the heat release, smoke and gas of EP/FePP nanocomposites are inhibited by CNTs-OLDH hybrid, and the PHRR, THR and SPR values of EP/4CNTs-OLDH nanocomposites decrease by 41.7%, 27.8% and 31.7%. The improved fire retardant performances and thermal stability are attributed to the excellent homogeneous dispersion, the network structure formed by the 3D hybrid in the matrix and the outstanding flame retardant effect of CNTs and OLDH.
RESUMO
The response rules of pressing pain on the back section in the Governor Vessel in patients with gastro-esophageal reflux disease (GERD) were studied to provide references for the diagnosis and treatment of GERD. Seventy-six cases of GERD were included into an observation group while 30 healthy volunteers were recruited into a control group. A mechanical measurement device of pressing pain that could measure the pain threshold was adapted to observe the pressing pain on the back section in the Governor Vessel in GERD patients and healthy volunteers. The test area is from spinous process of the 1st thoracic vertebra to that of the 12th thoracic vertebra (T1 -T12), including acupoints and non-acupoints on the Governor Vessel. As a result, in the observation group the pain threshold of T5-T7 spinous process clearance, which was the location of Shendao (GV 11), Lingtai (GV 10) and Zhiyang (GV 9), was lower than that in the control group (all P < 0.05). This result indicated that there was significant pressing pain in T5-T7 spinous process clearance in patients with GERD, which could be taken as an important auxiliary diagnosis and a new thinking method in the treatment of GERD with acupuncture.