Location:Home >> Research >> Research news
Contact Infomation

Name:ZHANG Xiujun

Tell:

Email:zhangxj@wbgcas.cn

Organization:Wuhan Botanical Garden

Related News

Research Proposes Comprehensive and High-performance Multi-task Deep Learning Framework: cmFormer

2024-04-26

The exponential progress in single-cell multi-omics technologies has led to the accumulation of extensive and diverse multi-omics datasets. Nevertheless, the integration of single-cell proteomics and transcriptomics (or epigenomics) data present considerable challenge for existing methods. Several Transformer-based models, such as Geneformer, have significantly transformed the paradigm of single-cell transcriptome analysis. However, these methods impose substantial demands on computational resources.

To tackle these challenges, Plant Bioinformatics Research Group of Wuhan Botanical Garden, Chinese Academy of Sciences, has developed a method called scmFormer based on the Transformer for integrating large-scale single-cell proteomics and transcriptomics data by multi-task transformer.

This study presented a comprehensive evaluation and made case studies of this method, the results revealed that scmFormer exhibited remarkable proficiency in harmonizing large-scale single-cell omics plus proteomics datasets at both the cell type and finer-scale cell levels with limited computer resource. scmFormer also possessed the capability to integrate multiple single-cell paired multimodal datasets, leading to a dual benefit of reduced highly cost and enhanced biological insights. Moreover, scmFormer shows an outstanding ability to eliminate technical disparities between different omics modalities while preserving the underlying biological information inherent in the data, encompassing both cell types and experimental conditions.

The successful applying of scmFormer to integrate two COVID-19 datasets comprising 1.48 million cells further testified the distinctive advantage of scmFormer in handling extensive datasets on regular laptops.

The research entitled "scmFormer integrates large-scale single-cell prote-omics and transcriptomics data by multi-task transformer” was published in the journal Advanced Science. XU Jing is the first author, and Professor ZHANG Xiujun is the corresponding author. This research was supported by the National Key Research and Development Plan, the National Natural Science Foundation and other projects.


Copyright 2002 - 2023 Wuhan Botanical Garden,Chinese Academy Of Sciences
Email: wbgoffice@wbgcas.cn     ICP: 05004779-1