●A preliminary work in which the graph neural network framework was used to encode/train heterogeneous biological data●Developed a pipeline in which some deep learning models were built and used to annotate cluster specific genes found from single cell RNAseq dataDeveloping a pipeline to predict interface regions and interface structures between two proteins which may form a protein complex by using 3D protein analysis tools and deep/machine learning methods.

●Analyzed high throughout sequencing data including RNAseq and TCRseq from an internal Phase III trial to support AACR/ASCO abstract submission and coming publication tasks●Using single cell data to find potential drug targets from CD4/CD8 T cells●Using deep learning software to analyze and integrate public single cell datasets●Analyze cold tumors or metastatic tumors to understand possible immune escape or resistant mechanisms●Did some preliminary analysis work about antibody engineering and design including antibody structure modeling, alanine mutation and MD-based free energy perturabtion.

As the project leader, presided over a Shanghai Natural Fund project which systematically studied on the translational medical significance of human long non-coding RNAsExplore the communication and interaction modes with research groups and develop differentiated and high-level bioinformatics services. Provide professional bioinformatics analysis support from the perspective of translational medicine research and clinical applicationWork with doctors/professors from hospitals in shanghai (ruijin, changzheng, Xinhua etc.) by finding probable causes of diseases of interest such as rare pediatric cancers from public data and in-house dataSummarize key findings from different subgroups of a 973 Program of lung cancer and mined something important which may be significantly correlated with disease mechanism of lung cancersDeveloped a new algorithm to identify significant changes of splicing events and applied it to explore whether PTEN regulate alternative splicing process. This work has been published listing as one of first authors (Nat Commun. 2018 Jun 19;9(1):2392).Participated in projects related to tumor microenvironment and drug resistance mechanism. As a co-author, some work has been published (Nat Commun. 2018 Oct 17;9(1):4315).Participated in projects related to innate immune cells and stem cells. As a co-author, some work has been published (J Allergy Clin Immunol. 2018 Apr 3. pii: S0091-6749(18)30453-6.).For the First time explored the biological functions of long non-coding RNA SNHG8 in cooperation with Fujian Cancer Hospital. This work has been published listing as one of first authors (Oncotarget. 2016 Dec 6;7(49):80990-81002). Moreover, this article has been included by GeneCards to decribe the functions of SNHG8.Participated in projects related to integration analysis of single cell RNA-seq, ATAC-seq/Chip-seq and RNA-seq to find driver genes to explain cell population heterogeneity.

Be familiar with strategies to find targets from genomic alterations including mutation, gene alternative splicing and gene fusions.Proactively communicate with scientists of both local and US to understand their needs and finally make deliveries reachable.Always query against some knowledge databases for drug R&D including Citeline, GOBIOM, Thomson Reuters Integrity, Thomson Reuters Partnering, KDP, PharmaPendium, GeneGo, Oncomine and GeneCards and literature databases including Pubmed, Scopus and EMBASE.Evaluations of NGS toolsCompare differences of gene expression calling between RNA-Seq platform and Affymetrix platform.Compare isoform inference tools for RNA-Seq data.Compare gene fusion tools for RNA-Seq data.Compare read mapping tools for both DNA-Seq data and RNA-Seq data.Process RNA-Seq/aCGH data of some cancer for translational medicine center of AstraZeneca in China to find potential cancer targets/biomarkers which might benefit AstraZeneca projects in discovery or clinical phases.Develop a pipeline/strategy to comprehensively characterize cancer specific isoform drivers using RNA-Seq datasets of TCGA.Construct an automate workflow for bacterial genome denovo assembly. This workflow was for a big project in the R&D center of AstraZeneca in US. This workflow includes sequencing QC module, duplicate removal module and read assembly module. It is written in Linux bash scripts and PERL scripts.