Nat Genetics . . . Zhang's/Liu Yu team worked to develop a non-coding mutation analysis method for individualized tumors for individual cases.

98% of the human genome is noncoding, and more than 80% of the noncoding genome contains DNA functional elements (such as promoters, enhancers, etc.), which have the function of gene transcription regulation [1].these non coding regions play an important role in the occurrence and treatment of many diseases, such as tumors, and have been paid more and more attention by researchers [2].in recent years, the accumulation of tumor genome-wide sequencing data has promoted the analysis of genomic variation [3].the current large-scale tumor genome research is mostly limited to the protein coding region of the genome. There are still many tumors that can not identify the driving gene, which indicates that the non coding region may carry key tumor driven variation.however, the progress of systematic research on non coding genome is slow, which is mainly due to the lack of understanding of the functional elements and regulatory molecular mechanisms of NCG.current analysis strategies are limited to the high recurrence of variation and rely on large sample size data analysis (such as pan genomics analysis).however, the multiple regulatory patterns in the three-dimensional spatial structure of transcriptional regulation (i.e., the variation of different DNA regulatory elements with a long linear distance has similar regulatory activity for target genes) further reduces the statistical efficiency of the analysis strategy, leading to the further increase of sample size dependence.the deep dependence on sample size greatly limits the analysis and understanding of non coding genome variation, and seriously hinders the application of non coding genome analysis in precision medicine.in view of this frontier scientific problem, on July 6, 2020, Liu Yu research group of Shanghai Children's Medical Center / National Children's Medical Center (Shanghai) affiliated to Shanghai Jiaotong University School of medicine and Zhang Jinghui's research group of Department of computational biology of St. Jude Children's research hospital, cooperated with other research teams in nature on July 6, 2020 Discovery of regulatory non coding variants in individual cancer genes by using cis-x was published in genetics In this paper, we developed a new computational analysis method cis-x for non coding genomic variation with transcriptional activation function, which broke through the dependence of sample size in noncoding genome variation analysis, and could carry out non coding variation analysis for tumor genome of a single case.cis-x adopts a bottom-up analysis strategy to analyze the allelic specific expression (ASE) and outlier high expression (out side high expression) of cis-x caused by non coding regulatory variation, Based on these two characteristics, we integrated the whole genome sequencing and transcriptome sequencing data of a single tumor sample to obtain the potential transcriptional activation target genes in the tumor genome, and then combined with the topologically associated domain (TAD) in the three-dimensional genome, we further screened the non coding variants with CIS activation function.this method can be used to analyze SNV / indel, copy number variation and chromosome rearrangement.it is worth noting that the research team has developed a new "balanced transcription" for ASE analysis, which is the core of cis-x, through years of accumulation of whole genome sequencing and transcriptome sequencing analysis The model can effectively model and correct the ASE signal characteristics and noise under different transcriptome sequencing depths, which significantly improves the ASE analysis efficiency and accuracy of cis-x in different sequencing depths.on the basis of previous work (Nature Genetics 2017 [4], nature 2018 [5]), the research team first selected children T-ALL to test the accuracy of cis-x.through individual non coding genomic analysis of 13 children with T-ALL admitted to Shanghai Children's Medical Center, 14 of the 15 non coding driven mutations were successfully detected by cis-x (the detection rate was 93.3%).in addition, cis-x analysis found a new somatic variation in the two patients. Through the formation of a new transcription factor YY1 binding site, the active enhancer was introduced into the mutation region, thus CIS activating the abnormal transcription of important proto oncogene TAL1 in T-ALL.this finding has been verified in vivo in a PDX model constructed by patient-derived tumor cells carrying mutations, which expands the understanding of the molecular mechanism of abnormal activation of the proto oncogene TAL1.in addition, cis-x analysis also found a novel candidate proto oncogene PRLR in T-ALL. the deletion of the upstream copy number of the PRLR gene leads to the structural changes of the TAD where the PRLR gene is located, and the abnormal transcription of PRLR is activated by enhancer hijacking. the team further verified the activation of PRLR by reanalyzing the T-ALL data of children in target project. genomic copy number variation is a common mutation type in tumor genome, and immune cell infiltration is common in many solid tumors. Both of them will interfere with the analysis of the CIS activation feature of gene ase. the researchers further selected children's neuroblastoma (target project) and adult melanoma data (TCGA project) with higher copy number variation and mutation load to verify the analysis efficiency of cis-x under these influences. the results showed that cis-x had the same high sensitivity in detecting non coding regulatory variation in tumor samples with genomic copy number amplification and immune cell infiltration, which could meet the requirements of different types of tumor analysis in children and adults, and has broad application prospects. in conclusion, the research team has developed a new personalized computational biology analysis tool cis-x for non coding regulatory variation. the preliminary application of cis-x in tumor genome analysis of children and adults shows that cis-x can effectively discover new noncoding variants and their regulated proto oncogenes in small sample size, supplement and expand the understanding of tumor genome, and open the possibility of application of non coding genome variation analysis in individualized precision medicine, and fill in the blank in this field. the first author and co-author of the paper is Dr. Liu Yu, the corresponding author is Dr. Zhang Jinhui, and the co-author is Dr. Li Chunliang, assistant professor of tumor cell biology department of St. Jude Children's research hospital, and Dr. Shen Shuhong, Department of Hematology, Shanghai Children's medical center. Dr. Liu Yu was engaged in post doctoral research in computational biology in the research group of Professor Zhang Jinhui of St. Jude Children's research hospital from 2014 to 2018. In 2018, he returned home full-time to join the Shanghai Children's Medical Center Affiliated to Shanghai Jiaotong University Medical School and served as the leader of the course. the research group of Liu Yu has long been committed to the research and development of analysis methods for children's tumor genomics, relying on the close cooperation of Hematology Oncology Department of Shanghai Children's medical center to explore the clinical transformation of genomic research. in recent years, they have successively published in the international authoritative journals Nature (2018), Nature Genetics (2020, 2017), cancer discovery (2018) and nature communications (2016) published important academic progress. The research group has long recruited co PI and postdoctoral researchers, and invited aspiring people with relevant research background in computational biology, genomics, tumor biology to join us liuyu@scmc.com.cn , yu.liu @ stju.edu.cn ）。 Code Project consortium. An integrated Encyclopedia of DNA elements in the human genome. Nature 489, 57-74 (2012). 2. Mansour, M. R. et al. Oncogene regulation. An oncogenic super enhancer formed through natural mutation of a non coding intergenic element. Science 346, 1373 – 1377 (2014).3. Rheinbay, E. et al. Analyses of non-coding somatic drivers in 2,658 cancer whole genomes. Nature 578, 102–111 (2020).4. Liu, Y. et al. The genomic landscape of pediatric and young adult T-lineage acute lymphoblastic leukemia. Nat. Genet. 49, 1211–1218 (2017).5. Ma, X.#, Liu, Y.# et al. Pan-cancer genome and transcriptome analyses of 1,699 paediatric leukaemias and solid tumours. Nature 555, 371–376 (2018).