Method and apparatus for deriving novel anticancer drugs using co-essentiality networks

Abstract

To provide a co-essentiality network-based anticancer drug derivation method that is more accurate than other network-based anticancer drug derivation methods.SOLUTION: A co-essentiality network of the present invention determines a cancer-specific anticancer drug, discovers drug-repurposing candidates, forms a highly clustered network module of a potential drug target, and predicts drug response in cancer cells more accurately than other molecular networks. Therefore, the present invention should be a valuable resource in precision oncology, and is expected to provide new treatment opportunities for cancer patients.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] This application relates to a method and apparatus for deriving novel anti-cancer drugs using co-essentiality networks. [Background technology]

[0002] Over the past few years, various anti-cancer drugs have been developed for various cancer types. However, the overall therapeutic efficacy of approved drugs is limited. Therefore, identifying targetable alternatives is an urgent challenge in anti-cancer treatment. Research into precision oncology requires various efforts, such as identifying driver genes and discovering drug targets for specific cancer types.

[0003] Network-based approaches are useful in precision oncology research for identifying robust anticancer targets or biomarkers linked to known disease genes because genes associated with disease phenotypes cluster together within the network. The present inventors recently identified peripheral chemotherapy and immunotherapy biomarkers by propagating therapeutic element associations from drug targets to their neighbors within a protein-protein interaction (PPI) network (Korea Patent Registration No. 10-2470937). Cheng et al. developed an in-silico framework for cancer drug repurposing using network modules derived from gene coexpression and PPI networks.

[0004] However, although the choice of network determines the performance of a network-based approach, it is still unknown which network is suitable for precision oncology. It has been confirmed that network topology is an important factor in improving disease gene identification. Huang et al. demonstrated that the difference between the minimum and maximum network performance is at least 24,000 times greater. Buphamalai et al. also reported that each network is relevant for specific tasks.

[0005] In addition, Chinese Patent No. 111128299 provides an invention for diagnosing colorectal cancer by creating a network using gene expression information, and Chinese Patent No. 110473591 provides an invention for creating a network using gene co-expression information. While there are cases where co-essentiality has been utilized, such as finding protein complexes using a co-essentiality network, there have been no cases where it has been used to derive new anticancer drugs (including repurposing existing drugs). [Prior art documents] [Patent documents]

[0006] [Patent Document 1] Korean Patent No. 10-2470937 [Patent Document 2] China Patent No. 111128299 [Patent Document 3] China Patent No. 110473591 [Non-patent literature]

[0007] [Non-Patent Document 1] Cheng F, Lu W, Liu C, Fang J, Hou Y, Handy DE, et al. A genome-wide positioning systems network algorithm for in silico drug repurposing. Nat Commun [Internet]. 2019 [cited 2020 Jul 19];10:1-14. Available from: https: / / doi.org / 10.1038 / s41467-019-10744-6 Summary of the Invention [Problem to be solved by the invention]

[0008] The problem to be solved by the present invention is to provide a co-essentiality network-based anticancer drug derivation method that is more accurate than other network-based anticancer drug derivation methods. The inventors designed and evaluated a network that best performs the intended task, rather than a uniform network. [Means for solving the problem]

[0009] The present application provides a method for deriving a new anticancer drug using a co-essentiality network by a computer device, comprising: (1) obtaining gene-genomic data, measuring the similarity between genes in the gene-genomic data, and creating a co-essentiality network; (2) extracting a cancer-related driver module from the co-essentiality network; (3) deriving the new anticancer drug using the cancer-related driver module.

[0010] The present application also provides a device for deriving a new anticancer drug using a co-essentiality network by a computer device, comprising: The present invention provides an apparatus including a data acquisition unit that acquires genetic genome data, a network creation unit that measures the similarity between genes in the genetic genome data and creates a co-essentiality network, a module extraction unit that extracts a cancer-related driver module from the co-essentiality network, and an anticancer drug derivation unit that uses the cancer-related driver module to derive the new anticancer drug. [Effects of the Invention]

[0011] The co-essentiality network of the present invention was able to identify cancer-specific anticancer drugs and discover drug-repurposing candidates. The co-essentiality links formed a highly clustered network module of potential drug targets. Furthermore, the co-essentiality network was able to predict drug responses in cancer cells more accurately than other molecular networks. Therefore, the present invention should be a valuable resource in precision oncology and is expected to provide new treatment opportunities for cancer patients. [Brief explanation of the drawings]

[0012] [Figure 1] (A) Schematic diagram of the construction and evaluation of the co-essentiality network, (B) abundance of co-essentiality links for KEGG pathways, (C) abundance of network links for 31 KEGG cancer-related pathways, and (D) relative modularity for 15 overlapping pathways among 31 CRPs and 41 co-essentiality-specific KEGG pathways. [Figure 2] The abundance of co-essentiality links for six types of genes is shown. [Figure 3](A) Modularity calculated for subnetworks of driver genes across 19 TCGA cancer types in four networks. (B)-(E) Subnetworks of LUSC driver genes in each network. (F) Schematic of identifying driver genes using co-essentiality networks. (G) Driver gene identification performance of the four networks. (H) Schematic of grouping TCGA patients using co-essentiality networks. (I) Results of patient grouping using the four networks and cancer drivers. [Figure 4] Empirical distribution of degree-controlled arbitrary node modularity for LUSC driver genes. [Figure 5] CGC cancer driver modularity of four types of networks is shown. [Figure 6] The clustering coefficients calculated for cancer driver genes in the four types of networks are shown. [Figure 7] In the co-essentiality network, the modular correlation between driver genes and the number of cell lines used in the network construction are shown. [Figure 8] The results show the performance of the co-essentiality network in identifying cancer driver genes compared with seven other PPI networks. [Figure 9] The cancer driver gene identification performance of the co-essentiality network was compared with a co-essentiality network different from the network of the present application and a gene interaction network. [Figure 10] (A) Schematic diagram of the Honet2 algorithm, and (B) a heatmap showing the performance ranking of 13 types of networks using MCC to identify driver genes. [Figure 11](A) Schematic diagram of the uKIN algorithm, and (B) a heatmap showing the performance ranking of identifying driver genes using ROAUC among 13 types of networks. [Figure 12] Survival plots for patient subgroups across 16 cancer types grouped as co-essentiality networks are shown. [Figure 13] Survival plots for patient subgroups of 16 cancer types grouped as PPI-BioGRID are shown. [Figure 14] Survival plots for patient subgroups across 16 cancer types grouped as co-expression networks are shown. [Figure 15] Survival plots for patient subgroups across 16 cancer types grouped as co-methylation networks are shown. [Figure 16] Survival plots of patient subgroups for 16 cancer types grouped by driver genes for each cancer type are shown. [Figure 17] (A) Subnetworks of the co-essentiality network used for signaling by the nonreceptor tyrosine kinase pathway, (B) Heatmap of LUSC gene expression within signaling by the nonreceptor tyrosine kinase pathway, (C) Survival plots of LUSC patient subgroups grouped by gene expression of signaling by the nonreceptor tyrosine kinase pathway, and (D) Survival plots of LUSC patient subgroups grouped by LUSC driver genes. [Figure 18](A) Schematic diagram of prioritizing drug targets using co-essentiality networks and the evaluation of each target; (B) NES by GSEA of four types of networks; (C) Subnetworks of five approved targets (CDKN1A, ATM, CRKL, SOX10, and RAF1) in SKCM and their driver genes in the nearest neighbors in the co-essentiality network; (D) Nearest neighbors of SOX10 in the four types of genes. [Figure 19] The results show the performance of the co-essentiality network in prioritizing approved anticancer targets compared with seven PPI networks. [Figure 20] This shows the results of comparing the performance of the co-essentiality network in prioritizing approved anti-cancer targets with a different co-essentiality network and a gene interaction network from the present application. [Figure 21] (A) Performance of the four networks in predicting drug response measured using the Spearman correlation coefficient between TC score and median log10(IC50). (B) Scatter plot of colorectal cancer drug response and TC score for the co-essentiality network and candidate drug data points. (C) Subnetworks of seven target genes of TAK-733 and COADREAD driver genes connected within the co-essentiality network. (D) Schematic diagram of drug RGE effect on COADREAD. (E) Scatter plot of drug RGE effect and TC score for COADREAD within the co-essentiality network. (F) Performance of the four networks in predicting COADREAD drug RGE effect. [Figure 22] The results show the drug response prediction performance of the co-essentiality network compared with seven types of PPI networks. [Figure 23] This shows the results of comparing the drug response prediction performance of the co-essentiality network with a different co-essentiality network from the present application. [Figure 24](A) Chord plot showing overall representation of potential anticancer drugs for 333 approved drugs across 17 cancer types; (B) Percentage TC scores for ixazomib citrate in four networks; (C) Coessentiality links between ixazomib citrate and target genes of LIHC driver genes; (D) Dose-response curves of ixazomib citrate for four LIHC cell lines (SNU-398, Huh7, SK-Hep-1, and HepG2); (E) Schematic of prolonged colony assay (left), showing the negative effect of ixazomib citrate on colony formation in four LIHC cell lines (SNU-398, Huh7, SK-Hep-1, and HepG2). [Figure 25] Links between ixazomib citrate targets and LIHC driver genes in the four networks are shown. [Figure 26] 1 shows a flowchart of the method according to the present application. [Figure 27] 1 shows a block diagram of an apparatus according to the present application. DETAILED DESCRIPTION OF THE INVENTION

[0013] Hereinafter, embodiments of the present application will be described in detail with reference to the accompanying drawings so that those skilled in the art can easily carry out the present application. However, the present application may be embodied in various different forms and is not limited to the embodiments described herein. In the drawings, parts that are not relevant to the description are omitted for clarity, and similar parts are designated by similar reference numerals throughout the specification.

[0014] Throughout this specification, when a member is said to be "on" another member, this includes not only when the member is in contact with the other member, but also when there is another member between the two members.

[0015] Throughout this specification, when a part is said to "comprise" a certain element, this means that it may further include other elements, but not to the exclusion of other elements, unless otherwise specified.

[0016] As used throughout the specification of this application, terms of degree such as "about," "substantially," etc., are used to mean a numerical value or a approximation of a numerical value when manufacturing and material tolerances inherent in the recited meaning are given, and are used to prevent unscrupulous infringers from unfairly taking advantage of disclosures in which precise or absolute numerical values ​​are recited to aid in the understanding of this application. As used throughout the specification of this application, the terms "steps to" or "steps of" do not mean "steps for."

[0017] Throughout the specification of this application, the term "combination(s) of these" contained in a Markush expression means a mixture or combination of one or more selected from the group of elements set forth in the Markush expression, and means including one or more selected from the group of elements.

[0018] Throughout the specification of this application, the phrase "A and / or B" means "A or B, or A and B."

[0019] Throughout this specification, the term "essential" refers to the degree of perturbation effect on a gene. For example, if a gene's removal at a particular level in a group of cells or a population negatively impacts the survival of the group, the gene can be said to be "essential" for the group.

[0020] Throughout this specification, the term "co-essentiality network" refers to a network composed of genes with similar knockout essentiality across various tumor cell lines. Although co-essentiality networks have already been used to predict the endowment of a gene, the benefits of using co-essentiality in drug discovery remain to be verified. For example, it has been reported that if two genes share similar essentiality information, they may have similar biological functions. Co-essentiality networks have been used to identify new gene possibilities and include genes in new clusters or metabolic pathways. However, attempts to use co-essentiality networks to identify alternative targets for many difficult-to-identify target proteins have been limited, despite the therapeutic opportunities offered by essentiality phenotypes.

[0021] Therefore, we evaluated various in-silico frameworks that could use co-essentiality networks to identify new anticancer drugs for specific cancer types and investigated the advantages of co-essentiality networks over existing molecular networks. As a result, co-essentiality networks were able to identify cancer-specific anticancer drugs and discover drug-repurposing candidates. Co-essentiality links formed highly clustered network modules of potential drug targets. Furthermore, co-essentiality networks predicted drug responses in cancer cells more accurately than other molecular networks. Therefore, co-essentiality networks should be a valuable resource in precision oncology and are expected to provide new therapeutic opportunities for cancer patients.

[0022] Throughout this specification, "drug-repurposing" is also referred to as "drug repurposing" and means repurposing a drug used to treat one disease to use it in the treatment of another disease.

[0023] Throughout this specification, the term "module" refers to a set of highly interrelated nodes within a network, and the term "driver module" refers to a module within a co-essentiality network that contains a cancer-related driver gene.

[0024] Throughout the present specification, the term "reversal gene expression" refers to a change in the gene expression pattern of a cell that is caused by the same factor as a drug, but in the opposite way.

[0025] Throughout this specification, "prioritization" refers to the process of prioritizing data according to priority.

[0026] Throughout this specification, "enrichment" means the increased expression of a gene for a phenotype.

[0027] Hereinafter, embodiments and examples of the present application will be described in detail with reference to the accompanying drawings, but the present application is not limited to these embodiments and examples and drawings.

[0028] Example 1. Data used in this application, network derivation process, etc. 1-1. Data and network information In the 13 networks used in this study, the nodes in the networks were converted to HUGO symbols, and only the nodes and links contained in the largest connected component were selected. The number of nodes and links in the final networks used in this study is shown in Table 1.

[0029] [Table 1]

[0030] 1-2. Gene essentiality data To create the co-essentiality network, the inventors used a dataset composed of genome-wide CRISPR screening data (Achilles project 20q2 in the dependency map (DepMap) project). Using this dataset and cell lines, growth data was obtained to examine how loss of function of some genes in a cell line affects the growth of the cell line. Here, the growth data is data on the essentiality of certain genes. The above data was used to create the co-essentiality network.

[0031] 1-3.Gene expression data To create the co-expression network, we used CCLE expression data quantified in RNA-seq files using the GTEx pipeline. The dataset contained gene expression data for 19,144 genes from 1,305 cell lines representing 34 different lineages. Of the 19,144 genes, 23 with expression values ​​of 0 across all cell lines were removed.

[0032] 1-4.Gene methylation data To construct the co-methylation network, we used CCLE DNA methylation reduced representation bisulfite sequencing data (promoter 1 kb upstream of TSS). The dataset consisted of methylation data for 21,337 loci covering 17,182 gene promoter regions in 843 cell lines. Due to the large number of missing values ​​in the data matrix, we retained only the 17,000 loci with methylation data for at least 644 cell lines and cell lines with loci with methylation data. Finally, we integrated 20,198 methylated loci from 805 cell lines into the network construction stage.

[0033] 1-5. Creating a network based on gene correlation In creating three networks, a co-essentiality network, a co-expression network, and a co-methylation network, from the datasets corresponding to each network, the inventors measured the similarity between two genes in terms of essentiality, expression, and methylation, and used this similarity as the link weight of the network. After correcting the data, if both genes had a link weight of 0, the link was excluded. The above process may consist of the following three steps:

[0034] (1) For missing values ​​in the dataset, we performed k-nearest neighbor (KNN) imputation with k = 10 using the impyute Python module. (2) To measure the correlation between genes, we calculated the Pearson correlation coefficient (PCC) for every gene pair in the dataset, and used the absolute value of the PCC to capture both directions of correlation between genes. (3) The absolute PCC values ​​are subjected to the context likelihood relatedness (CLR) algorithm, which performs adaptive background correction to remove spurious correlations and indirect effects. For example, the PCC value between genes i and j, r ij teeth,

number

[0035] [Table 2]

[0036] 1-6.PPI Network Eight protein-protein interaction (PPI) networks were used: BioGRID, BioPlex, GPSnet, HURI, Inbiomap, iRefIndex, Pathway Commons, and STRING. Link weights were set to 1.0 for all PPI networks.

[0037] 1-7.BioGRID The BIORID interactome was downloaded from https: / / thebiogrid.org / using BIOGRID-4.1.190. The interaction between both human proteins was used.

[0038] 1-8.BioPlex The following BioPlex interactors were downloaded: BioPlex 3.0 Interactions (293T Cells) and BioPlex HCT116 v.1.0 (HCT116 cells) (https: / / bioplex.hms.harvard.edu / interactions.php under release BioPlex 3.0). We created a network by combining both interactors.

[0039] 1-9.GPSnet We use the GPSnet interactor previously constructed by Cheng et al., which combines 15 commonly used databases, including a wealth of experimental evidence and in-house systematic human protein-protein interactors. The interactor is publicly available at https: / / github.com / ChengF-Lab / GPSnet / tree / master / Data_mat and the "Net_PPI.mat" file. GPSnet is originally an in-silico framework for drug repurposing, and in this application, the PPI network used in this framework is referred to as GPSnet.

[0040] 1-10.HURI HURI interactors were downloaded from http: / / www.interactome-atlas.org / and the “HuRI.tsv” file.

[0041] 1-11.InBioMap InBioMap interactors were downloaded from http: / / www.intomics.com / inbio / map and the file “InBio_Map_core_2016_09_12”.

[0042] 1-12.iRefIndex iRefIndex version 13.0 was downloaded from the web interface or Interaction Reference Index repository (iRefWeb, https: / / wodaklab.org / iRefWeb / search / index). We used the search options "single organism interaction," "Homo Sapiens," "experimental," and "physical."

[0043] 1-13.Pathway Commons PathwayCommons was downloaded from http: / / www.pathwaycommons.org / archives / PC2 / v12 / and the “PathwayCommons12.All.hgnc.txt” file.

[0044] 1-14.STRING STRING interaction v11.0 was downloaded from https: / / string-db.org / . To avoid co-citation information, we removed text-mining driver genes from all links and recalculated the STRING confidence scores. Links with a confidence score of over 700 were used as highly reliable PPIs.

[0045] 1-15. Gene interaction network To compare the synthetic-lethal relationship-based gene interaction network with the co-essentiality network, we used a clinically relevant synthetic lethality network created using the "identification of clinically relevant synthetic lethality (ISLE)" approach, which was downloaded from https: / / github.com / jooslee / ISLE / tree / main / networks and the "ISLE_clinical_SL_network_FDR_0.2.cys" file.

[0046] 1-16. Cutted gene set enrichment analysis (GSEA) for co-essentiality networks We calculated the enrichment of co-essentiality links within six introduced gene sets: KEGG [1], REACTOM [2], and Gene Ontology annotations [3] [GO: BP (Biological Process), MF (Molecular Function), and CC (Cellular Components)] (sources: molecular signatures database (MsigDB) [4] and human core protein complex from CORUM [5]).

[0047] In co-essentiality, gene pairs were grouped into bins of 10,000 pairs, scored by link weights, and abundance was calculated as the proportion of pairs with the same biological module annotation. Abundance was also calculated by taking into account the probability of finding gene pairs in the same biological module without information from co-essentiality links. Similar to the method used by Lee et al. [6], we replaced the denominator in the above probability from the number of co-essentiality links to the number of all possible gene pairs within a given bin.

[0048] 1-17. Modularity calculation The inventors used two measures of modularity according to the following functions:

number

[0049] Because both modularity measurement methods can affect the degree of centrality of the query node, we applied normalization to eliminate precision bias. Similar to the method of Guney et al. [7], we created a reference modularity distribution corresponding to the predicted modularity of 100 arbitrary gene groups, matched to the size and number of links (degree) of the genes of interest in the 100 networks. We then normalized the modularity as the z-score of the measured gene modularity calculated from the reference modularity distribution of the arbitrary group.

[0050] 1-18. Network enrichment for cancer-related pathways (CRPs) Of the 186 KEGG pathways in MsigDB, 31 pathways related to "Pathways in cancer" (Pathway ID: hsa05200) and pathways in "Cancer: specific types" were used as CRPs. The 31 KEGG pathway list is shown in Table 3.

[0051] [Table 3a] [Table 3b] [Table 3c] [Table 3d] [Table 3e] [Table 3f] [Table 3g] [Table 3h] [Table 3i] [Table 3j] [Table 3k] [Table 3l]

[0052] For the 186 KEGG pathways, we calculated relative modularity by converting the calculated modularity using cohesiveness to give percentage scores in four types of networks: co-essentiality networks, PPI networks (BioGRID), co-expression networks, and co-methylation networks. We then assigned a network type to each of the 186 KEGG pathways based on which network showed the highest relative modularity, thereby defining a network-specific pathway. The relative modularity values ​​are shown in Table 3.

[0053] Finally, for each network, we constructed four 2 × 2 contingency tables, designated network-specific CRPs, non-network-specific CRPs, network-specific non-CRPs, and non-network-specific non-CRPs. In the contingency tables, we calculated the odds ratio of network enrichment for CRP and determined statistical enrichment by calculating p-values ​​from Fisher's exact test. The contingency tables are shown in Figure 1C.

[0054] 1-19. Driver genes for each cancer type (1) Bailey et al. [8] We searched for driver genes for each cancer type in the Bayley study, which reported 299 driver genes across 33 cancer types. For modularity analysis, we selected 19 cancer types with 10 or more reported driver genes included in the co-essentiality network. The cancer types mentioned above are: bladder urothelial carcinoma (BLCA), breast invasive carcinoma (BRCA), colon adenocarcinoma (COAD), lymphoid neoplasm diffuse large B-cell lymphoma (DLBC), head and neck squamous cell carcinoma (HNSC), kidney chromophobe (KICH), kidney renal clear cell carcinoma (KIRC), acute myeloid leukemia (LAML), brain lower grade glioma. (LGG), liver hepatocellular carcinoma (LIHC), lung adenocarcinoma (LUAD), lung squamous cell carcinoma (LUSC), Ovarian serous cystadenocarcinoma (OV), pancreatic adenocarcinoma (PAAD), prostate adenocarcinoma (PRAD), skin cutaneous melanoma (SKCM), stomach adenocarcinoma (STAD), Thyroid carcinoma (THCA), and Uterine corpus endometrial carcinoma (UCEC).

[0055] (2) Cancer Gene Census (CGC) To further verify the high modularity of the driver gene co-essentiality network, we used an experimentally validated set of cancer driver genes from the Cancer Gene Census (CGC) database. For cancer type analysis, we manually mapped the CGC tier 1 driver genes to the tumor type information for TCGA cancer types. A total of 470 driver genes were mapped to 22 TCGA cancer types. For modularity analysis, we selected 19 cancer types with 10 or more reported driver genes included in the co-essentiality network. The cancer types are: bladder urothelial carcinoma (BLCA), breast invasive carcinoma (BRCA), colon adenocarcinoma (COAD), lymphoid neoplasm diffuse large B-cell lymphoma (DLBC), head and neck squamous cell carcinoma (HNSC), kidney chromophobe (KICH), kidney renal clear cell carcinoma (KIRC), acute myeloid leukemia (LAML), brain lower grade glioma. (LGG), liver hepatocellular carcinoma (LIHC), lung adenocarcinoma (LUAD), lung squamous cell carcinoma (LUSC), Ovarian serous cystadenocarcinoma (OV), pancreatic adenocarcinoma (PAAD), prostate adenocarcinoma (PRAD), skin cutaneous melanoma (SKCM), stomach adenocarcinoma (STAD), Thyroid carcinoma (THCA), and Uterine corpus endometrial carcinoma (UCEC).

[0056] Thanks for watching the snowstorm. TCGA_ACC : TP53 TCGA_GBM : ACVR1 , ATRX TCGA_ESCA: KDM6A, LRP1B, SFRP4, SOX2 TCGA_BLCA: DROSHA, ERBB3, FGFR3, HRAS, KDM6A, LRP1B, MDM4, MTOR, NOTCH1, NOTCH2, STAG2, TERT, TSC1 TCGA_LGG: ATM, AXIN1, DDX3X, EPAS1, KDM6A, KMT2C, KMT2D, NBN, PALB2, PMS2: PTCH1, QKI, SUFU, TERT TCGA_HNSC: BCORL1, CTCF, ERBB3, FAT1, FAT4, FES, MET, MTOR, NFE2L2, NOTCH1, PTK6, PTPRT, TGFBR2, TP63, TSC2 TCGA_LIHC: APC, ARID1B, ARID2, AXIN1, AXIN2, CASP8, CTNNB1, DNAJB1, FAT4, HNF1A, IL6ST, PRKACA, PTPN13, SMAD2, TERT TCGA_SKCM: BLM, BRAF, DDB2, ERBB3, ERCC2, ERCC3, ERCC4, ERCC5, FAS, LATS1, NOTCH1, POLE, PTCH1, RECQL4, SMO, STAT5B, TBX3, TERT, XPA, XPC TCGA_STAD: ACVR2A, ATR, AXIN2, BCL9L, BRAF, CDH1, ERBB2, ERBB3, ERBB4, FGFR2, GRIN2A, PIK3CA, POLE, PTK6, PTPN13, PTPRT, RHOA, SFRP4, UBR5, ZFHX3 TCGA_PAAD: ACVR2A, AKT2, APC, ATRX, BRAF, BRCA2, CDKN2A, DAXX, EP300, FAT1, FAT4, GNAS, HIF1A, KRAS, MAP2K4, MEN1, PREX2, RNF43, SMAD4, SND1, STK11 TCGA_PRAD: ACSL3, AR, AXIN1, BRAF, CANT1, DDX5, ELK4, ERG, ETV1, ETV4, ETV5, FOXA1, HERPUD1, HNRNPA2B1, KLF6, KLK2, NCOR2, NDRG1, PTEN, RAF1, SALL4, SLC45A3, SPOP, TMPRSS2, ZFHX3 TCGA_UCEC: ATR, AXIN1, BARD1, BCL9L, CHD4, CTCF, CUX1, FBXW7, FGFR2, HIF1A, MAX, MED12, MLH1, MSH2, MSH6, MTOR, NUTM2B, NUTM2D, PMS2, POLE, PTEN, RAD21, RAD51B, SPOP, SRC, SUZ12, YWHAE, ZFHX3, TCGA_THCA, BRAF, CCDC6, CDC73, CDKN1B, DICER1, ERC1, GOLGA5, HMGA1, HOOK3, KRAS, KTN1, MEN1, MTOR, NCOA4, NRAS, NTRK1, PAX8, PCM1, PPARG, PRKAR1A, RET, STRN, TFG,TPM3, TPR, TRIM27, TRIM33, TSHR, ZNF331, TCGA_KICH, AMER1, ARID1A, BAP1, CLTC, DICER1, DROSHA, FH, FLCN, GPC3, HIF1A, KDM6A, KMT2D, MET, MYCN, NF2, NONO, PALB2, PRCC, PTK6, SFPQ, SIX1, SMARCB1, TFE3, TFEB, TMEM127, TP53, TSC1, TSC2, VHL, WT1 TCGA_OV: AKT1, AKT2, ARID1A, ARID1B, ATR, BARD1, BRAF, BRCA1, BRCA2, CCNE1, CDK12, CTNNB1, ERBB2, FES, FOXL2, GOPC, LRP1B, MAPK1, MLH1, MSH2, MSH6, PIK3R1, PMS2, PPM1D, PPP2R1A, PTK6, RNF43, ROS1, SMARCA4 STK11 TCGA_KIRC: AMER1, ARID1A, BAP1, CLTC, DICER1, DROSHA, FH, FLCN, GPC3, HIF1A, KDM6A, KMT2D, MET, MTOR, MYCN, NF2, NONO, PALB2, PRCC, PTK6, SFPQ, SIX1, SMARCB1, TFE3, TFEB, TMEM127, TP53, TSC1, TSC2, VHL, WT1 TCGA_BRCA: AKT1, APOBEC3B, ARID1A, ARID1B, BAP1, BARD1, BRCA1, BRCA2, BRIP1, CASP8, CCND1, CDH1, CDKN1B, CHEK2, CTCF, EP300, ERBB2, ESR1, ETV6, FOXA1, GATA3, IRS4, KEAP1, MAP2K4, MAP3K1, MAP3K13, NCOR1, NOTCH1, NTRK3, PALB2, PBRM1, PIK3CA, POLQ, PPM1D, RB1, SALL4, SMARCD1, TBX3, TP53 TCGA_COAD: AKT1, APC, AXIN1, AXIN2, B2M, BAX, BCL9L, BRAF, CTNNB1, CUX1, EIF3E, EP300, ERBB3, FBXW7, GRIN2A, HIF1A, KRAS, MAP2K1, MAP2K4, MAX, MDM2, MLH1, MSH2, MSH6, MUTYH, PIK3CA, PIK3R1, PMS2, POLD1, POLE, PTPRK, PTPRT, QKI, RAD21, RSPO2, RSPO3, SALL4, SFRP4, SMAD2, SMAD3, SMAD4, SRC, TBL1XR1, TCF7L2, TGFBR2, TP53, UBR5, WDCP TCGA_LUAD: AKT1, ALK, BAP1, BRAF, CCDC6, CD74, DDR2, DICER1, DROSHA, EGFR, EML4, ERBB2, ERBB4, EZR, FGFR2, GRIN2A, HIF1A, HIP1, KDR, KEAP1, KIF5B, KRAS, LRIG3, MAP2K1, MAP2K2, MYCL, NFE2L2, NKX2-1, NOTCH1, NRG1, PIK3CB, PTPN13, PTPRT, RAD21, RB1, RBM10, RET, ROS1, SDC4, SLC34A2, SMARCA4, SOX2, STK11, STRN, TFG, TP53, TP63, TPM3, TPR TCGA_LUSC: AKT1, ALK, BAP1, BRAF, CCDC6, CD74, DDR2, DICER1, DROSHA, EGFR, EML4, ERBB2, ERBB4, EZR, FGFR2, GRIN2A, HIF1A, HIP1, KDR, KEAP1, KIF5B, KRAS, LRIG3, MAP2K1, MAP2K2, MYCL, NFE2L2, NKX2-1, NOTCH1, NRG1, PIK3CB, PTPN13, PTPRT, RAD21, RB1, RBM10, RET, ROS1, SDC4, SLC34A2, SMARCA4, SOX2, STK11, STRN, TFG, TP53, TP63, TPM3, TPR TCGA_DLBC: ALK, ARHGAP26, ASXL1, ATIC, ATM, ATR, B2M, BCL10, BCL2, BCL6, BCL7A, BIRC3, BIRC3, BLM, BTK, CALR, CARD11, CARS, CBL, CCND2, CCND3, CD274, CD79A, CD79B, CDKN2C, CEBPA, CIITA, CLTC, CLTCL1, CNTRL, CNTRL, CREBBP, CSF3R, CUX1, DDX6, EIF4A2, EP300, ETV6, EZH2, FAS, FAT4, FBXO11, FCRL4, FGFR1, FGFR1, FGFR1OP, FGFR1OP, FGFR3, FGFR3, HIST1H4I, HSP90AA1, HSP90AB1, IGH, IGH, IGK, IGL, IKBKB, IKZF1, IL2, IL21R, IRF4, ITK, JAK2, KDM6A, KDSR, LEF1, MAF, MAFB, MALT1, MECOM, MN1, MPL, MSN, MUC1, MYC, MYD88, MYH9, NBN, NFKB2, NIN, NOTCH2, NPM1, NRAS, NSD2, PAFAH1B2, PAX5, PCM1, PDCD1LG2, PDE4DIP, PDGFRB, PIM1, POU2AF1, PRDM1, PRDM16, PRF1, PTPN11, REL, RHOA, RHOH, RMI2, RNF213, SBDS, SETBP1, SF3B1, SH2B3, SOCS1, SPEN, SRSF2, SRSF3, STAG2, SYK, SYK, TBL1XR1, TENT5C, TET2, TFG, TNFAIP3, TNFRSF14, TNFRSF17, TP63, TPM3, TPM4, U2AF1, UBR5, WAS, ZMYM2, ZMYM2, ZRSR2 TCGA_LAML: ABI1, ABL1, ABL2, AFDN, AFF1, AFF3, AFF4, ARHGAP26, ARHGAP26, ARHGEF12, ARNT, ASXL1, ASXL1, ATM, ATR, BCL11A, BCL11B, BCL2, BCL3, BCL6, BCL9, BCOR, BCORL1, BCR, BIRC3, BIRC3, BLM, BRCA2, BRIP1, BTG1, BTK, CALR, CBFA2T3, CBFB, CBL, CBL, CBLB, CBLC, CCDC6, CCND1, CCND2, CCND3, CDK6, CDKN2C, CDX2, CEBPA, CEBPA, CNOT3, CNTRL, CREBBP, CRLF2, CSF3R, CSF3R, CUX1, CUX1, DDX10, DDX3X, DEK, DNM2, DNMT3A, ELF4, ELL, EP300, EPS15, ERG, ETNK1, ETV6, EWSR1, FANCA, FANCC, FANCD2, FANCE, FANCF, FANCG, FAT1, FBXW7, FCGR2B, FGFR1, FGFR1OP, FGFR3, FLT3, FOXO3, FOXO4, FOXP1, FSTL3, FUS, GAS7, GATA1, GATA2, GPHN, HIP1, HLF, HOXA11, HOXA13, HOXA9, HOXC11, HOXC13, HOXD11, HOXD13, IGH, IGH, IKZF1, IL7R, IRF4, IRS4, JAK1, JAK2, JAK2, JAK3, KAT6A, KAT6B, KDM5A, KDM6A, KDM6A, KIT, KMT2A, KNL1, KRAS, LASP1, LCK, LEF1, LMO1, LMO2, LPP, LRP1B, LYL1, MAF, MAFB, MAPK1, MECOM, MECOM, MLF1, MLLT1, MLLT10, MLLT11, MLLT3, MLLT6, MN1, MN1, MPL, MRTFA, MSI2, MTCP1, MYC, MYH11, NCOA2, NFKBIE, NIN, NOTCH1, NPM1NRAS, NRAS, NSD1, NSD2, NSD3, NT5C2, NUMA1, NUP214, NUP98, OLIG2, P2RY8, PALB2, PAX5, PBX1, PCM1, PCM1, PDE4DIP, PDGFRB, PDGFRB, PER1, PHF6, PICALM, PML, POT1, PRDM16, PRDM16, PRF1, PRRX1, PSIP1, PTPN11, PTPN11, PTPRC, RABEP1, RAD21, RAP1GDS1, RARA, RBM15, RPL10, RPL22, RPL5, RPN1, RUNX1, RUNX1T1, SBDS, SBDS, SET, SETBP1, SETBP1, SF3B1, SH2B3, SH2B3, SH3GL1, SRSF2, SRSF2, STAG2, STAT3, STAT5B, STIL, SYK, TAF15, TAL1, TAL2, TBL1XR1, TCF3, TCL1A, TENT5C, TET1, TET2, TLX1, TLX3, TOP1, TRA, TRB, TRD, TRIM24, TRIP11, U2AF1, U2AF1, XPO1, ZBTB16, ZMYM2, ZNF384, ZNF521, ZRSR2, ZRSR2,

[0057] 1-20. Network propagation using driver genes for each cancer type To prioritize genes in the network based on their distance from the driver gene, we performed network propagation using the PageRank algorithm in the NetworkX Python module. Among the 19 cancer types, we performed network propagation on the 17 cancer types whose co-essentiality networks showed the highest modularity. For each cancer type, we assigned a value of 1 to the driver gene and a value of 0 to the other genes in the network, which were then used as input values ​​for the personalization parameters in the PageRank algorithm. All other PageRank algorithm parameters were adjusted to the base value (damping factor = 0.85).

[0058] 1-21. Grouping similar patients and comparing survival rates by patient group To identify driver modules that can differentiate survival rates between patient groups, we used the network propagation scores of genes in the network to identify biological pathways associated with driver genes. To calculate the propagation driver genes of genes included in each pathway, we performed gene set enrichment analysis (GSEA) using the GSEApy Python module. After performing GSEA on the network propagation driver genes, we selected pathways that were particularly enriched in genes with high network propagation driver genes (FDR < 0.001 and NES > 0) as driver modules.

[0059] The patient transcripts and clinical data tables used in this study were downloaded using the TCGAbiolink R package. To preprocess the gene expression data from TCGA patients, we used gene length-corrected, trimmed mean M-values ​​calculated using the edgeR R package and calculated gene expression levels using normalized read counts. For statistical significance, we selected cancer types containing at least 100 samples from 17 cancer types with the highest driver gene modularity co-essentiality network. This resulted in the selection of 16 cancer types containing 7,259 tumor samples.

[0060] The inventors divided TCGA patients into two groups based on the gene expression levels of driver modules. For each patient, a single-sample gene set abundance analysis (ssGSEA) was performed on the gene expression of selected driver modules. The top 50% of patients based on NES were defined as the up-regulated group, and the bottom 50% were defined as the down-regulated group. To determine whether there was a difference in survival rate between the up-regulated and down-regulated groups, a log-rank test was used to obtain statistical significance. To compare the maximum ability of patient subclassification between networks, the driver module of the patient group with the lowest p-value in the log-rank test was selected, and the p-value is shown in Figure 3I.

[0061] 1-22. Drug-Gene Association Data We used drug-gene association data obtained by combining 18 sources in PanDrugs. This source includes expert-compiled data and drug-gene association data collected through experimental drug screening. The original data in PanDrugs were provided by the original authors. We used two drug-gene association typologies, "direct target" and "biomarker," to identify drug targets. Finally, we considered 43,909 drug-target associations across 9,090 drugs.

[0062] 1-23. Prioritization Performance of FDA-Approved Anticancer Targets To determine whether the network propagation results could prioritize the targets of FDA-approved drugs in each cancer type, we performed GSEA on the propagation results of driver genes for that cancer type. To do so, we collected a list of 273 FDA-approved drugs across 17 cancer types from TCGA (https: / / www.cancer.gov / about-cancer / treatment / drugs) and the PanDrugs database. For each cancer type, a list of genes with drug-target relationships between approved drugs and genes was used for GSEA. The network's ability to prioritize approved anti-cancer targets was measured by the NES value of the GSEA results. The FDA-approved anti-cancer drugs and their targets used in this analysis are listed below.

[0063] [Approved Drugs] BLCA ERDAFITINIB|ENFORTUMAB VEDOTIN-EJFV|MITOMYCIN|VALSTAR|ATEZOLIZUMAB| KEYTRUDA|METHOTREXATE SODIUM|BALVERSA|OPDIVO|BAVENCIO|VALRUBICIN|IMFINZI| PEMBROLIZUMAB|TECENTRIQ|PADCEV|METHOTREXATE|DOXORUBICIN HYDROCHLORIDE| DURVALUMAB|CISPLATIN|THIOTEPA|NIVOLUMAB|JELMYTO|AVELUMAB BRCA EXEMESTANE|FEMARA|ERIBULIN|ADO-TRASTUZUMAB EMTANSINE|EPIRUBICIN| LYNPARZA|IBRANCE|TRASTUZUMAB AND HYALURONIDASE-OYSK|ATEZOLIZUMAB|TAXOTERE| METHOTREXATE SODIUM|PERJETA|PHESGO|ZOLADEX|TALZENNA|KISQALI|KADCYLA| METHOTREXATE|ELLENCE|TAMOXIFEN CITRATE|TUKYSA|ABRAXANE|VINBLASTINE| NERATINIB MALEATE|FARESTON|PIQRAY|VERZENIO|AFINITOR|HERCEPTIN|TOREMIFENE| ANASTROZOLE|PACLITAXEL|TAMOXIFEN|AFINITOR DISPERZ|AREDIA|GOSERELIN ACETATE| TREXALL|PERTUZUMAB|MEGESTROL ACETATE|ENHERTU|SACITUZUMAB GOVITECAN-HZIY|5-FU| OLAPARIB|PALBOCICLIB|THIOTEPA|LAPATINIB DITOSYLATE|AROMASIN| PACLITAXEL ALBUMIN-STABILIZED NANOPARTICLE FORMULATION|TALAZOPARIB TOSYLATE| RUCAPARIB (CAMSYLATE)|IXEMPRA|HERCEPTIN HYLECTA|FAM-TRASTUZUMAB DERUXTECAN-NXKI| ALPELISIB|ARIMIDEX|MITOXANTRONE|EPIRUBICIN HYDROCHLORIDE|FULVESTRANT|GEMZAR| XELODA|GEMCITABINE HYDROCHLORIDE|RUCAPARIB|TECENTRIQ|TUCATINIB|IXABEPILONE| PAMIDRONATE DISODIUM|RIBOCICLIB|NERLYNX|VINBLASTINE SULFATE|EVEROLIMUS|FASLODEX| TOREMIFENE CITRATE|ABEMACICLIB|TRODELVY|HALAVEN|DOCETAXEL|CAPECITABINE|ERIBULIN MESYLATE|LAPATINIB|CYCLOPHOSPHAMIDE|MITOXANTRONE HYDROCHLORIDE|MELPHALAN| DOXORUBICIN HYDROCHLORIDE|LETROZOLE|PERTUZUMAB, TRASTUZUMAB, AND HYALURONIDASE- ZZXF|NERATINIB|RIBOCICLIB SUCCINATE|TRASTUZUMAB|FLUOROURACIL INJECTION|TYKERB CESC PEMBROLIZUMAB|MVASI|AVASTIN|KEYTRUDA|BEVACIZUMAB|TOPOTECAN|TOPOTECAN HYDROCHLORIDE|BLEOMYCIN SULFATE|HYCAMTIN COADREAD YERVOY|AVASTIN|LONSURF|ZIV-AFLIBERCEPT|TIPIRACIL HYDROCHLORIDE| TRIFLURIDINE AND TIPIRACIL HYDROCHLORIDE|KEYTRUDA|IRINOTECAN HYDROCHLORIDE| IPILIMUMAB|VECTIBIX|IRINOTECAN|ERBITUX|OPDIVO|CAPECITABINE|PANITUMUMAB|5-FU| BEVACIZUMAB|XELODA|CAMPTOSAR|PEMBROLIZUMAB|MITOMYCIN C|TIPIRACIL| LEUCOVORIN CALCIUM|OXALIPLATIN|CYRAMZA|MVASI|CETUXIMAB|RAMUCIRUMAB|ELOXATIN| ZALTRAP|NIVOLUMAB|STIVARGA|REGORAFENIB|FLUOROURACIL INJECTION ESCA KEYTRUDA|RAMUCIRUMAB|OPDIVO|TRASTUZUMAB GBM MVASI|AVASTIN|AFINITOR DISPERZ|GLIADEL WAFER|PROCARBAZINE|TEMODAR| CARMUSTINE|AFINITOR|TEMOZOLOMIDE|EVEROLIMUS|CARMUSTINE IMPLANT|BICNU| BEVACIZUMAB|LOMUSTINE HNSC CETUXIMAB|KEYTRUDA|TREXALL|HYDREA|TAXOTERE|PEMBROLIZUMAB|NIVOLUMAB| HYDROXYUREA|METHOTREXATE|ERBITUX|OPDIVO|DOCETAXEL|BLEOMYCIN SULFATE KIRC AVASTIN|AFINITOR|IL-2|SORAFENIB|INTERLEUKIN-2|PROLEUKIN|NEXAVAR| EVEROLIMUS|PAZOPANIB HYDROCHLORIDE|KEYTRUDA|IPILIMUMAB|SUNITINIB| SUNITINIB MALATE|AFINITOR DISPERZ|OPDIVO|BAVENCIO|TEMSIROLIMUS|YERVOY| RELATED RESOURCES|TORISEL|BEVACIZUMAB|AXITINIB|PEMBROLIZUMAB|MITOMYCIN| PAZOPANIB|ALDESLEUKIN|JELMYTO|SUTENT|MVASI|CABOMETYX|CABOZANTINIB-S-MALATE| LENVIMA|NIVOLUMAB|SORAFENIB TOSYLATE|VOTRIENT|INLYTA|AVELUMAB| LENVATINIB MESYLATE LGG MVASI|AVASTIN|AFINITOR DISPERZ|GLIADEL WAFER|PROCARBAZINE|TEMODAR| CARMUSTINE|AFINITOR|TEMOZOLOMIDE|EVEROLIMUS|CARMUSTINE IMPLANT|BICNU| BEVACIZUMAB|LOMUSTINE LIHC PEMIGATINIB|CYRAMZA|REGORAFENIB|ATEZOLIZUMAB|CABOMETYX|KEYTRUDA| BEVACIZUMAB|STIVARGA|CABOZANTINIB-S-MALATE|LENVIMA|PEMBROLIZUMAB|AVASTIN| TECENTRIQ|SORAFENIB TOSYLATE|FLOXURIDINE|RAMUCIRUMAB|NIVOLUMAB|NEXAVAR|OPDIVO| PEMAZYRE|LENVATINIB MESYLATE LUAD CYRAMZA|IRESSA|AVASTIN|TARCEVA|OSIMERTINIB MESYLATE|AFINITOR| ALECTINIB HYDROCHLORIDE|PORTRAZZA|SELPERCATINIB|VINORELBINE| TOPOTECAN HYDROCHLORIDE|PORFIMER|METHOTREXATE SODIUM|ENTRECTINIB|IPILIMUMAB| CERITINIB|IMFINZI|GILOTRIF|TAFINLAR|ALIMTA|ERLOTINIB HYDROCHLORIDE|XALKORI| EVEROLIMUS|ALECENSA|GEMCITABINE|MEKINIST|PACLITAXEL|TAXOTERE|TRAMETINIB| CARBOPLATIN|P”RAPL’T|AFINITOR DISPERZ|DOCETAXEL|DABRAFENIB MESYLATE|OPDIVO| TAGRISSO|BEVACIZUMAB|PACLITAXEL ALBUMIN-STABILIZED NANOPARTICLE FORMULATION| YERVOY|BRIGATINIB|GEMZAR|MUSTARGEN|TREXALL|VIZIMPRO|VINORELBINE TARTRATE| ALUNBRIG|GEMCITABINE HYDROCHLORIDE|TOPOTECAN|DACOMITINIB|ROZLYTREK|TECENTRIQ| LORBRENA|METHOTREXATE|PEMETREXED|PEMETREXED DISODIUM|ATEZOLIZUMAB|ABRAXANE| PEMBROLIZUMAB|MVASI|DOXORUBICIN HYDROCHLORIDE|PARAPLATIN|LORLATINIB|RETEVMO| RAMUCIRUMAB|MECHLORETHAMINE HYDROCHLORIDE|GEFITINIB|AFATINIB| CAPMATINIB HYDROCHLORIDE|DURVALUMAB|NIVOLUMAB|CRIZOTINIB|KEYTRUDA|TABRECTA| ZYKADIA|AFATINIB DIMALEATE|NECITUMUMAB|ALECTINIB|ERLOTINIB LUSC CYRAMZA|IRESSA|AVASTIN|TARCEVA|OSIMERTINIB MESYLATE|AFINITOR| ALECTINIB HYDROCHLORIDE|PORTRAZZA|SELPERCATINIB|VINORELBINE| TOPOTECAN HYDROCHLORIDE|PORFIMER|METHOTREXATE SODIUM|ENTRECTINIB|IPILIMUMAB| CERITINIB|IMFINZI|GILOTRIF|TAFINLAR|ALIMTA|ERLOTINIB HYDROCHLORIDE|XALKORI| EVEROLIMUS|ALECENSA|GEMCITABINE|MEKINIST|PACLITAXEL|TAXOTERE|TRAMETINIB|CARBOPLATIN|PARAPLAT|AFINITOR DISPERZ|DOCETAXEL|DABRAFENIB MESYLATE|OPDIVO|TAGRISSO| BEVACIZUMAB|PACLITAXEL ALBUMIN-STABILIZED NANOPARTICLE FORMULATION|YERVOY| BRIGATINIB|GEMZAR|MUSTARGEN|TREXALL|VIZIMPRO|VINORELBINE TARTRATE|ALUNBRIG| GEMCITABINE HYDROCHLORIDE|TOPOTECAN|DACOMITINIB|ROZLYTREK|TECENTRIQ|LORBRENA| METHOTREXATE|PEMETREXED|PEMETREXED DISODIUM|ATEZOLIZUMAB|ABRAXANE|PEMBROLIZUMAB| MVASI|DOXORUBICIN HYDROCHLORIDE|PARAPLATIN|LORLATINIB|RETEVMO|RAMUCIRUMAB| MECHLORETHAMINE HYDROCHLORIDE|GEFITINIB|AFATINIB|CAPMATINIB HYDROCHLORIDE| DURVALUMAB|NIVOLUMAB|CRIZOTINIB|KEYTRUDA|TABRECTA|ZYKADIA|AFATINIB DIMALEATE| NECITUMUMAB|ALECTINIB|ERLOTINIB PRAD PROVENGE|RADIUM 223 DICHLORIDE|RUBRACA|ELIGARD|XTANDI|FIRMAGON| FLUTAMIDE|ENZALUTAMIDE|NILUTAMIDE|LYNPARZA|ERLEADA|DEGARELIX ACETATE| CABAZITAXEL|MITOXANTRONE|ABIRATERONE|NILANDRON|JEVTANA|LEUPROLIDE ACETATE| BICALUTAMIDE|LUPRON DEPOT|RUCAPARIB CAMSYLATE|GOSERELIN|GOSERELIN ACETATE| XOFIGO|MITOXANTRONE HYDROCHLORIDE|LUPRON|DEGARELIX|MELPHALAN|ESTRAMUSTINE| TAXOTERE|ZOLADEX|OLAPARIB|DAROLUTAMIDE|ESTRAMUSTINE PHOSPHATE SODIUM|DOCETAXEL| NUBEQA|CASODEX|ABIRATERONE ACETATE|APALUTAMIDE|SIPULEUCEL-T| ESTRAMUSTINE PHOSPHATE|ZYTIGA SKCM SONIDEGIB|FLUOROURACIL-TOPICAL|COBIMETINIB|VINBLASTINE SULFATE|ERIVEDGE| VISMODEGIB|PACLITAXEL|VORINOSTAT|TRAMETINIB|ODOMZO|ALDARA|COBIMETINIB FUMARATE| DACARBAZINE|DABRAFENIB MESYLATE|VEMURAFENIB|1218778-77-8|ERISMODEGIB|DABRAFENIB| 1187431-43-1|5-FU|VINBLASTINE|EFUDEX|ALITRETINOIN|IMIQUIMOD|AVELUMAB STAD CYRAMZA|DOXORUBICIN HYDROCHLORIDE|TRIFLURIDINE AND TIPIRACIL HYDROCHLORIDE|PEMBROLIZUMAB|EVEROLIMUS|5-FU|RAMUCIRUMAB|TRASTUZUMAB| TAXOTERE|HERCEPTIN|MITOMYCIN|SUNITINIB|SUNITINIB MALATE|LANREOTIDE|DOCETAXEL| MITOMYCIN C|KEYTRUDA|LONSURF|REGORAFENIB|FLUOROURACIL INJECTION UCEC HYDROXYPROGESTERONE CAPROATE|DACTINOMYCIN|VINBLASTINE|METHOTREXATE SODIUM|LENVIMA|PEMBROLIZUMAB|MEGESTROL ACETATE|KEYTRUDA|METHOTREXATE|VINBLASTINE SULFATE|LENVATINIB MESYLATE UCSHYDROXYPROGESTERONE CAPROATE|DACTINOMYCIN|VINBLASTINE|METHOTREXATE SODIUM|LENVIMA|PEMBROLIZUMAB|MEGESTROL ACETATE|KEYTRUDA|METHOTREXATE|VINBLASTINE SULFATE|LENVATINIB MESYLATE [Target of approved drugs] BLCA ERDAFITINIB|ENFORTUMAB VEDOTIN-EJFV|MITOMYCIN|VALSTAR|ATEZOLIZUMAB| KEYTRUDA|METHOTREXATE SODIUM|BALVERSA|OPDIVO|BAVENCIO|VALRUBICIN|IMFINZI| PEMBROLIZUMAB|TECENTRIQ|PADCEV|METHOTREXATE|DOXORUBICIN HYDROCHLORIDE| DURVALUMAB|CISPLATIN|THIOTEPA|NIVOLUMAB|JELMYTO|AVELUMAB FKBP15|SALL3|NQO1| DUSP28|B2M|CD226|ATP9B|S100A12|GPR35|FGFR4|C18orf63|HDAC4|TAT|NHLRC1|SMIM21| CCDC102B|NTRK1|FGFR3|ANKMY1|NDUFA10|ZNF236|FAM132B|SCLY|CDH7|AC016757.3|XPC| ALO”5|RA’P1|CLCN6|PASK|FAM69C|PER2|SOCS6|PMS2|FBXO15|FGFR2|CNDP1|CNDP2|KRAS| ATIC|RBM44|CYP2B6|CD3EAP|ZNF516|ZADH2|KIT|PTEN|GALR1|NR1I2|NETO1|FOLH1|TMEM43| SLC19A1|CDKN1B|FGFR1|ANO7|MSH2|SLC31A1|AC110619.2|XRCC3|RP11-162A12.2| UBE2F-SCLY|RP11-861L17.3|MLH1|ASXL1|STK25|RP11-321M21.3|JAK2|JAK3|JAK1|CX3CL1| TRAF3IP1|RP11-723G8.2|PRR21|TOP2A|DSEL|CTLA4|FARP2|KLHL30|TSHZ1|PPP1R7|ESPNL| ASB1|AFP|IL1RN|TP53|ESR1|CD274|UBE2F|TIMM21|BAIAP3|CCND1|OTOS|HES6|TMX3|CBLN2| AC062017.1|RNPEPL1|POLE|RP11-41O4.1|ITIH1|AGXT|ITIH3|GLS|FLT3|KCNA1|RTTN|TPM3| CDH19|ICAM3|S100A8|LRRFIP1|KIF1A|BIRC5|SNED1|PDCD1|ACVR2A|ALDH3A1|ARVCF|MTHFR| MECP2|TPMT|COMT|BAX|GTSCR1|ADA|LINC00908|LRP2|GALNT14|ALK|GSTM1|GSTM4|CAPN10| IL2|C2orf54|ZNF407|SLAMF1|RP11-17M16.1|TOP2B|DHFR|ERCC2|MSH3|EGFR|MSH6|FCGR3B| SLCO1B1|AC093802.1|AC104809.3|ALDH1A1|OR6B2|OR6B3|HLA-DRA|AC079612.1|TWIST2| NRAS|RB1|DOK6|AQP12B|AQP12A|SEPT2|MBP|KRT20|SULT2A1|NR3C1|HDLBP|HSPB2| RP11-94B19.4|RP11-169F17.1|GPC1|ILKAP|WRAP53|HGF|CYB5A|MTERFD2|E2F1|NOTCH1| TYMS|GSTP1|KDR|MYEOV2|BRAF|ABCC4|PPP1R13L BRCA EXEMESTANE|FEMARA|ERIBULIN|ADO-TRASTUZUMAB EMTANSINE|EPIRUBICIN| LYNPARZA|IBRANCE|TRASTUZUMAB AND HYALURONIDASE-OYSK|ATEZOLIZUMAB|TAXOTERE| METHOTREXATE SODIUM|PERJETA|PHESGO|ZOLADEX|TALZENNA|KISQALI|KADCYLA| METHOTREXATE|ELLENCE|TAMOXIFEN CITRATE|TUKYSA|ABRAXANE|VINBLASTINE| NERATINIB MALEATE|FARESTON|PIQRAY|VERZENIO|AFINITOR|HERCEPTIN|TOREMIFENE| ANASTROZOLE|PACLITAXEL|TAMOXIFEN|AFINITOR DI”PERZ’AREDIA|GOSERELIN ACETATE| TREXALL|PERTUZUMAB|MEGESTROL ACETATE|ENHERTU|SACITUZUMAB GOVITECAN-HZIY|5-FU| OLAPARIB|PALBOCICLIB|THIOTEPA|LAPATINIB DITOSYLATE|AROMASIN|PACLITAXEL ALBUMIN-STABILIZED NANOPARTICLE FORMULATION|TALAZOPARIB TOSYLATE|RUCAPARIB (CAMSYLATE)|IXEMPRA|HERCEPTIN HYLECTA|FAM-TRASTUZUMAB DERUXTECAN-NXKI|ALPELISIB| ARIMIDEX|MITOXANTRONE|EPIRUBICIN HYDROCHLORIDE|FULVESTRANT|GEMZAR|XELODA| GEMCITABINE HYDROCHLORIDE|RUCAPARIB|TECENTRIQ|TUCATINIB|IXABEPILONE| PAMIDRONATE DISODIUM|RIBOCICLIB|NERLYNX|VINBLASTINE SULFATE|EVEROLIMUS|FASLODEX| TOREMIFENE CITRATE|ABEMACICLIB|TRODELVY|HALAVEN|DOCETAXEL|CAPECITABINE|ERIBULIN MESYLATE|LAPATINIB|CYCLOPHOSPHAMIDE|MITOXANTRONE HYDROCHLORIDE|MELPHALAN| DOXORUBICIN HYDROCHLORIDE|LETROZOLE|PERTUZUMAB, TRASTUZUMAB, AND HYALURONIDASE- ZZXF|NERATINIB|RIBOCICLIB SUCCINATE|TRASTUZUMAB|FLUOROURACIL INJECTION| TYKERBSTK11|PCSK6|DUSP28|CD226|SLC50A1|TAT|PIK3CA|PIK3CB|PIK3CD|PIK3CG|IRAK2| SCLY|CBLC|IL2RA|IFNG|CAMKV|PI”3C2B’PIK3C2A|PIK3C2G|MUC16|CYP3A4|CYP3A5|CNDP1| CNDP2|F5|POU2AF1|CD3EAP|T”BB4B’CCND3|CCND2|PTEN|TUBE1|DSTYK|CDKN1B|CDKN1A| ALDH3A1|EDNRB|SNORD68|RP11-162A12.2|RP11-321M21.3|JAK2|NF2|NF1|ZNF516|DSEL| ALPK2|DST|BLM|PPARD|CHST3|RTTN|LRMDA|PSIP1|CSNK1A1|TOP1|ITGAL|DCST2|HES6|SMAD4| RET|ITPR2|XIAP|TAF15|FH|GLS|TSC2|TSC1|NCOA3|NCOA1|PRAME|AFP|CKS1B|NR2F2|SPARC| S100A8|TFPI|NAT2|LTK|SRPK2|CYP2C9|SLCO1B1|F2|ALK|MAP2K1|KRTCAP2|PBRM1|ERBB2| TOP2B|MAP2K5|FLCN|DHFR|FLAD1|EGFR|BAX|CTNNB1|OR6B2|OR6B3|HLA-DRA|MYOD1|DPYD| CDK12|TWIST2|CXCR2|PGR|PGP|FOLH1|NOS1|ITGB2|AC110619.2|RP11-169F17.1|HGF| MTERFD2|EIF4E2|E2F1|BNC2|PRCC|HEXB|CYB5A|PPP1R13L|LIF|NQO2|NQO1|FCGR1A|PIK3C3| TUBA3E|TUBA3D|AD”MTS2'SIX4|ROR2|NDUFA10|CDH1|CXCR7|CDH7|CDKN2B|AC093802.1| HLA-DRB1|RAMP1|FAM69C|KLK3|ARID1A|C1QC|C1QB|C1QA|CCND1|NETO1|APC|ACTB|FOXP3| NOS3|CNTN5|SMARCB1|ATP7A|MYCN|RP11-861L17.3|CYP2D6|CCNE1|COL18A1|STK25|ITK| TUBD1|TRAF3IP1|PRR21|IGF2|FARP2|TUBB2A|TUBB2B|TUBB|PCDH15|GAS6|EPHX1|PDGFRA| HDAC2|HDAC4|ZNF236|ATM|CYP4B1|TIGD1|ATR|RP11-41O4.1|FLT3|CYP1B1|POLD1|PSEN1| TPM3|RP11-201K10.3|FANCC|AC104809.3|FANCA|LRRFIP1|FANCI|FBXW7|NTF3|TET2|MED24| DLL1|MTOR|GTSCR1|LINC00908|NPAT|KIF1A|SLC22A16|IRS1|CDK2|CDK4|CDK6|SLAMF1|TGFB1| POLA1|MSH2|MSH3|MSH6|MCL1|BGLAP|AC079612.1|RB1|SERPINE1|ETS2|ABCB1|AURKA|ABCB4| SULT2A1|PRKCA|FCGR2B|FCGR2C|FCGR2A|CES1|ABCC4|CES2|TUBA4A|CTNNA2|TOPBP1|NOTCH1| PARP3|HLA-DQA1|TUBG1|TYMS|GSTP1|MYEOV2|ABCC3|MMP2|TGM2|FBXO15|GPER1|GPR35| PPP1R7|MAP3K2|DNMT1|RNPEPL1|BRDT|JAK3|BRCA1|ERBB3|BRCA2|CETP|SHC1|ERBB4|IDH2| IDH1|CLCN6|PASK|PER2|RPL13|BRD4|CTSS|FGFR2|FGFR3|FGFR1|ATIC|RBM44|PLD1|SYK| CXCL8|GALR1|FKBP1A|ALPK1|GRIA3|HRAS|RRAS2|GAPDH|CRTC3|XRCC1|XRCC3|XRCC4|SEPT2| UBE2F-SCLY|PPM1D|GLDN|FOS|JUN|PPARGC1A|RRM2B|NFE2|CX3CL1|CHEK2|CHEK1|CTLA4| TUBB8|SULT1C4|TUBB6|TSHZ1|TUBB3|PARP2|TUBB1|SFPQ|ASB1|ESPNL|TP53|GNAQ|ESR1| CD274|PALB2|CYP19A1|UBE2F|HERC2|EIF4”BP1|'TOS|AC062017.1|SNED1|KCNA1|SLC10A2| JMJD6|ANKMY1|CD80|CBLN2|ICAM3|RRM2|DPM3|ERCC2|KDR|SRC|PRDX4|ACACA|EFNA4|EFNA3| EFNA1|PIP|SALL3|ESR2|CYP1A2|RPSA|GSTM1|TNFRSF1A|AGXT|IL15|CAPN10|IL5|IL2| C2orf54|ZNF407|RIPK4|CDH19|BIRC5|FCGR3A|BIRC2|FCGR3B|MAP2|SLCO1B3|DDIT3|MAP4| UGT1A5|ALDH1A1|FAS|MLLT6|AMHR2|PTHLH|NRG1|MBP|KRT20|ANXA1|HSPB2|ZBTB7B|MTR| ILKAP|CYP2A6|WRAP53|PARP1|EPCAM|CASP3|DRD4|DRD2|EBP|DCST1|MAPT|C1S|C1R|ABCC6| ABCC1|BCL2|PTGS2|GSR|FAM132B|AKT1|AKT2|AKT3|ATP9B|BDNF|S100A12|PBXIP1|C18orf63| CCDC102B|NTRK1|TLE3|CDKN2A|SFTPA1|EDN1|MYC|MYB|AC016757.3|SOD2|ALOX5|MGMT|PMS2| SMIM21|KRAS|FDPS|SOCS6|POLE|ZADH2|NR1I2|YSK4|SLC19A1|VDR|STMN1|TRAF2|CSF2|CLTC| ARID4B|CHD1|PTH|TUBA1C|TUBA1B|TUBA1A|PIK3R6|PIK3R5|PIK3R4|PIK3R3|PIK3R2|PIK3R1| RP11-723G8.2|TOP2A|TRH|KLHL30|STAG2|TRIM46|FGF2|IL1RN|ABCG2|MPL|SIK2|HOXD13| PRKCZ|EPHA4|EPHA3|GNRHR|ABL1|CGA|CACNB2|CYP2B6|MAPK3|FASN|PDCD4|TBXA2R|PYGO2| MTHFR|HDLBP|ARFGEF1|ADA|ADK|RASGRF2|MET|PRKACG|TIMM21|RP11-17M16.1|SPG7|KALRN| TFF3|VHL|LENEP|ANO7|ADAM15|RRM1|DSCAM|OPLAH|NRAS|MLH1|DOK6|AQP12B|AQP12A| CYP2C19|LDHA|NR3C1|RP11-94B19.4|PLAT|GPC1|GGT1|TMX3|SDHC|ABCC10|PMVK|ASXL1| PTPRC|BRAF|GNA11|BAP1 CESC PEMBROLIZUMAB|MVASI|AVASTIN|KEYTRUDA|BEVACIZUMAB|TOPOTECAN|TOPOTECAN HYDROCHLORIDE|BLEOMYCIN SULFATE|HYCAMTIN NRAS|HRAS|MSH3|FCGR3A|MSH2| FCGR3B|VHL|POLE|COL18A1|RB1|FCGR1A|ATRX|BDNF|HTRA1|HLA-DRA|TYMS|DPYD|PIK3CA| THBS2|MLH1|CXCR2|JAK2|NF2|NF1|JAK1|ANXA11|GGH|TOP1|PDCD1|ERBB2|PTEN|MTHFR|EGFR| IDH1|TOP1MT|FCGR2B|FCGR2C|FCGR2A|VEGFA|MSH6|VEGFC|VEGFB|MUC16|PMS2|B2M|SHMT1| KRAS|ARMS2|TP53|ALK|CXCL8|C1QC|C1QB|C1QA|KIT|MET|BRAF|MMP9|C1R|CD274|MMP2| SLC19A1 COADREAD YERVOY|AVASTIN|LONSURF|ZIV-AFLIBERCEPT|TIPIRACIL HYDROCHLORIDE| TRIFLURIDINE AND TIPIRACIL HYDROCHLORIDE|KEYTRUDA|IRINOTECAN HYDROCHLORIDE| IPILIMUMAB|VECTIBIX|IRINOTECAN|ERBITUX|OPDIVO|CAPECITABINE|PANITUMUMAB|5-FU| BEVACIZUMAB|XELODA|CAMPTOSAR|PEMBROLIZUMAB|MITOMYCIN C|TIPIRACIL| LEUCOVORIN CALCIUM|OXALIPLATIN|CYRAMZA|MVASI|CETUXIMAB|RAMUCIRUMAB|ELOXATIN| ZALTRAP|NIVOLUMAB|STIVARGA|REGORAFENIB|FLUOROURACIL INJECTION PTGS2|NT5E|AKT1| B2M|ARAF|FCGR1A|LGALS2|BDNF|EGF|TAT|PIK3CA|NTRK1|PTP4A3|CDH1|JAK2|FGFR1|BRCA1| ERBB3|BRCA2|ERBB4|IDH1|BGLAP|MGMT|CXCL10|PMS2|FGFR2|SHMT1|KRAS|CXCL8|C1QC|C1QB| C1QA|KIT|PTEN|APC|KLC3|SLC19A1|CDKN1A|MSH2|BCHE|XRCC1|ATP7A|ATP7B|RUNX3|THBS2| VPS37A|XDH|ISG15|NF2|NF1|JAK1|ANXA11|CTLA4|FRK|VEGFA|VEGFC|VEGFB|TP53|DDR2| CD274|PALB2|TOP1|VWF|KRT18|PDGFRB|PDGFRA|SMAD4|ATM|EPHA2|RET|POLE|MAPK10|MAPK11| ABL1|IL1A|FLT4|RAF1|HTRA1|FLT1|AREG|TEK|FANCC|GGH|KDR|FBXW7|MTHFR|FCGR3A|TOP1MT| CGB5|AGAP1|AC064874.1|MSH6|ATRX|TJP2|ARMS2|ALK|GSTM1|MET|EREG|PARD3B|RAD50| PDCD1|ERBB2|DHFR|ERCC1|ERCC2|MSH3|EGFR|BAX|FCGR3B|VHL|MAP2K1|CDKN2A|UGT1A1| BIRC5|HLA-DRA|TYMSOS|DPYD|NRAS|TYMP|MLH1|CXCR2|ABCB1|ABCB4|NRG1|PIGF|COL18A1| FCGR2B|FCGR2C|FCGR2A|CES1|CES2|CYP7B1|HRAS|E2F1|DKK1|TYMS|GSTP1|BRAF|MMP9|C1S| C1R|MMP2|ABCC1|BCL2 ESCA KEYTRUDA|RAMUCIRUMAB|OPDIVO|TRASTUZUMAB ERBB2|PDGFRA|BIRC5|FCGR3A| FCGR3B|AKT2|FCGR1A|MET|PIK3CA|PGR|KDR|ANXA1|ERBB3|IFNG|EGFR|FCGR2B|FCGR2C| FCGR2A|C1R|HGF|EPCAM|KRAS|ESR1|ARID1A|C1QC|C1QB|C1QA|PTEN|BRAF|C1S|APC GBM MVASI|AVASTIN|AFINITOR DISPERZ|GLIADEL WAFER|PROCARBAZINE|TEMODAR| CARMUSTINE|AFINITOR|TEMOZOLOMIDE|EVEROLIMUS|CARMUSTINE IMPLANT|BICNU| BEVACIZUMAB|LOMUSTINE BRCA1|GSR|SYNE1|STMN4|THBS2|HRAS|ATM|EGFR|PRL|RET|C1QB| CTNNB1|VHL|AKT2|AKT3|COL18A1|CSF2|FCGR1A|MMP9|MGMT|KIT|TSC2|HTRA1|TSC1|H2AFX| C1R|DPYD|PIK3CA|ATR|NRAS|RB1|CXCR2|AKT1|NF2|NF1|ANXA11|MYC|GGH|ERBB2|FLCN|BRCA2| STAG2|MTHFR|FCGR3A|IDH1|GNA11|FCGR2B|FCGR2C|PLAU|FCGR2A|MTOR|MSH6|VEGFC|VEGFB| ATRX|FKBP1A|ADK|SHMT1|KRAS|ARMS2|TP53|GNAQ|E2F1|ESR1|CXCL8|MPL|G6PD|C1QC|NOTCH1| C1QA|STK11|PTEN|MAOA|FCGR3B|BRAF|BAX|PBRM1|VEGFA|BAP1|SLC19A1|MMP2|MET HNSC CETUXIMAB|KEYTRUDA|TREXALL|HYDREA|TAXOTERE|PEMBROLIZUMAB|NIVOLUMAB| HYDROXYUREA|METHOTREXATE|ERBITUX|OPDIVO|DOCETAXEL|BLEOMYCIN SULFATESALL3| STK11|FBXO15|DUSP28|AKT1|B2M|CD226|RB1|FCGR1A|BDNF|S100A12|GPR35|EGF|C18orf63|TAT|PPP1R7|SMIM21|CCDC102B|PIK3CA|DNMT1|NTRK1|SMAD4|NDUFA10|PTP4A3|ZNF236|FAM132B|RRM2B|SCLY|CDH7|ERBB2|ERBB3|AC016757.3|AQP12A|ERBB4|ALOX5|RAMP1|AC062017.1|FAM69C|PER2|KLK3|RPL13|BRD4|MUC16|CYP3A4|CYP3A5|FGFR2|CNDP1|CNDP2|KRAS|ATIC|RBM44|CCND1|CXCL8|C1QC|C1QB|C1QA|KIT|PTEN|GALR1|NETO1|SULT2A1|SLC19A1|DSEL|CDKN1B|HRAS|ANO7|SNED1|ZADH2|XRCC1|AC110619.2|ARID4B|ATP7A|XRCC4|RP11-162A12.2|UBE2F-SCLY|RP11-861L17.3|NR1I2|PASK|NT5E|VPS37A|STK25|RP11-321M21.3|JAK2|NF2|JAK1|TRAF3IP1|RP11-723G8.2|PRR21|IGF2|FARP2|KLHL30|SULT1C4|TSHZ1|TUBB3|TUBB1|BLM|PPARD|ASB1|EIF4E|CHST3|ESPNL|AFP|C2orf54|IL1RN|TP53|DNTT|GPC1|CD274|ATP9B|TUBB|UBE2F|TIMM21|OTOS|AREG|HES6|GAS6|EPHX1|PMS2|IC4|TLE3|CBLN2|CYP4B1|RET|RNPEPL1|POLE|RP11-41O4.1|FH|GLS|FLT3|KCNA1|SLC10A2|ANKMY1|RTTN|TPM3|CDH19|ICAM3|S100A8|LRRFIP1|KIF1A|AC093802.1|DUSP1|MSH2|KDR|FBXW7|MSH3|MTHFR|EGFR|HDLBP|MSH6|GTSCR1|ADA|LINC00908|CP|TJP2|SLCO1B1|ALK|GSTM1|AGXT|MAP2K1|CAPN10|MET|IL2|EREG|ZNF407|PDCD1|SLAMF1|RRM2|RP11-17M16.1|RRM1|SPG7|DHFR|BIRC5|PRDX4|FCGR3A|BAX|FCGR3B|MAP2|SLCO1B3|ZNF516|MCL1|MAP4|XRCC3|OR6B2|OR6B3|HLA-DRA|AC104809.3|DPYD|AC079612.1|TWIST2|NRAS|MLH1|SNORD68|ABCB1|DOK6|AQP12B|PGP|NAT2|NRG1|SEPT2|MBP|FOLH1|NR3C1|HSPB2|RP11-94B19.4|RP11-169F17.1|FCGR2B|FCGR2C|ABL1|FCGR2A|ABCC4|SOCS6|ILKAP|CYP2A6|TMX3|HGF|CYB5A|MTERFD2|E2F1|PTH|NOTCH1|TYMS|GST”1|MY’OV2|BRAF|MAPT|C1S|C1R|ABCC6|BCL2. KIRC AVASTIN|AFINITOR|IL-2|SORAFENIB|INTERLEUKIN-2|PROLEUKIN|NEXAVAR| EVEROLIMUS|PAZOPANIB HYDROCHLORIDE|KEYTRUDA|IPILIMUMAB|SUNITINIB|SUNITINIB MALATE|AFINITOR DISPERZ|OPDIVO|BAVENCIO|TEMSIROLIMUS|YERVOY|RELATED RESOURCES| TORISEL|BEVACIZUMAB|AXITINIB|PEMBROLIZUMAB|MITOMYCIN|PAZOPANIB|ALDESLEUKIN| JELMYTO|SUTENT|MVASI|CABOMETYX|CABOZANTINIB-S-MALATE|LENVIMA|NIVOLUMAB|SORAFENIB TOSYLATE|VOTRIENT|INLYTA|AVELUMAB|LENVATINIB MESYLATE STK11|PLK4|AKT1|AKT2|AKT3| ARAF|MLH1|FCGR1A|PMS2|B2M|PIK3CA|PHC1|FGFR1|BRCA1|FLCN|BRCA2|IL2RA|IDH1|KRAS| FGFR2|FGFR3|SHMT1|ITK|CXCL8|C1QC|C1QB|C1QA|KIT|PTEN|XRN1|FKBP1A|SLC19A1|HLA-B| HRAS|HIF1A|CSF1|ASXL1|RAF1|CCNE1|THBS2|FLT3|JAK2|NF2|NF1|JAK1|ANXA11|TSC1|CTLA4| ERP27|VEGFA|VEGFC|VEGFB|FGF3|FGF1|TP53|GNAQ|ESR1|DDR2|CD274|MPL|HMOX1|LCK| PDGFRB|PDGFRA|MKI67|ATM|RET|POLE|ABL1|FLT4|TSC2|HTRA1|FLT1|CSF1R|GGH|KDR|FBXW7| ACVR2A|MTHFR|FCGR3A|MTOR|SH2B3|ARMS2|IL2RB|ALK|IL2RG|MET|PBRM1|CDK8|PDCD1|ERBB2| MSH2|MSH3|EGFR|MSH6|FCGR3B|CTNNB1|VHL|UGT1A1|HLA-DRA|CDK19|DPYD|NRAS|RB1|CXCR2| PGR|AURKB|PTPRB|PDGFA|PDGFB|PDGFC|PDGFD|COL18A1|GNA11|FCGR2B|FCGR2C|FCGR2A|CA9| NOTCH1|U2AF1|BRAF|MMP9|C1R|BAP1|MMP2 LGG MVASI|AVASTIN|AFINITOR DISPERZ|GLIADEL WAFER|PROCARBAZINE|TEMODAR| CARMUSTINE|AFINITOR|TEMOZOLOMIDE|EVEROLIMUS|CARMUSTINE IMPLANT|BICNU| BEVACIZUMAB|LOMUSTINE BRCA1|GSR|SYNE1|STMN4|THBS2|HRAS|ATM|EGFR|PRL|RET|C1QB| CTNNB1|VHL|AKT2|AKT3|COL18A1|CSF2|FCGR1A|MMP9|MGMT|KIT|TSC2|HTRA1|TSC1|H2AFX| C1R|DPYD|PIK3CA|ATR|NRAS|RB1|CXCR2|AKT1|NF2|NF1|ANXA11|MYC|GGH|ERBB2|FLCN|BRCA2| STAG2|MTHFR|FCGR3A|IDH1|GNA11|FCGR2B|FCGR2C|PLAU|FCGR2A|MTOR|MSH6|VEGFC|VEGFB| ATRX|FKBP1A|ADK|SHMT1|KRAS|ARMS2|TP53|GNAQ|E2F1|ESR1|CXCL8|MPL|G6PD|C1QC|NOTCH1| C1QA|STK11|PTEN|MAOA|FCGR3B|BRAF|BAX|PBRM1|VEGFA|BAP1|SLC19A1|MMP2|MET LIHC PEMIGATINIB|CYRAMZA|REGORAFENIB|ATEZOLIZUMAB|CABOMETYX|KEYTRUDA| BEVACIZUMAB|STIVARGA|CABOZANTINIB-S-MALATE|LENVIMA|PEMBROLIZUMAB|AVASTIN| TECENTRIQ|SORAFENIB TOSYLATE|FLOXURIDINE|RAMUCIRUMAB|NIVOLUMAB|NEXAVAR|OPDIVO| PEMAZYRE|LENVATINIB MESYLATE PDGFRB|PDGFRA|NRAS|HRAS|MSH3|FCGR3A|EPHA2|RET| FCGR3B|PTEN|VHL|POLE|COL18A1|MAPK11|FCGR1A|BCL2|ABL1|FLT4|RAF1|HTRA1|FLT1|JAK3| HLA-DRA|TEK|MLH1|DPYD|PIK3CA|DNMT1|THBS2|NTRK1|ARAF|FLT3|CXCR2|JAK2|NF2|NF1| JAK1|ANXA11|MSH2|TYMS|FGFR1|GGH|KDR|FBXW7|ERBB2|CTLA4|CD274|APC|MTHFR|EGFR|IDH1| FCGR2B|FCGR2C|FCGR2A|FRK|CX3CL1|ENO2|VEGFA|MSH6|VEGFC|VEGFB|PMS2|B2M|FGFR2| SHMT1|KRAS|ARMS2|TP53|ALK|DDR2|CXCL8|C1QC|C1QB|C1QA|KIT|MET|FAS|BRAF|MMP9|C1R| PDCD1|MMP2|SLC19A1 LUAD CYRAMZA|IRESSA|AVASTIN|TARCEVA|OSIMERTINIB MESYLATE|AFINITOR|ALECTINIB HYDROCHLORIDE|PORTRAZZA|SELPERCATINIB|VINORELBINE|TOPOTECAN HYDROCHLORIDE| PORFIMER|METHOTREXATE SODIUM|ENTRECTINIB|IPILIMUMAB|CERITINIB|IMFINZI|GILOTRIF| TAFINLAR|ALIMTA|ERLOTINIB HYDROCHLORIDE|XALKORI|EVEROLIMUS|ALECENSA|GEMCITABINE| MEKINIST|PACLITAXEL|TAXOTERE|TRAMETINIB|CARBOPLATIN|PARAPLAT|AFINITOR DISPERZ| DOCETAXEL|DABRAFENIB MESYLATE|OPDIVO|TAGRISSO|BEVACIZUMAB|PACLITAXEL ALBUMIN-STABILIZED NANOPARTICLE FORMULATION|YERVOY|BRIGATINIB|GEMZAR|MUSTARGEN| TREXALL|VIZIMPRO|VINORELBINE TARTRATE|ALUNBRIG|GEMCITABINE HYDROCHLORIDE| TOPOTECAN|DACOMITINIB|ROZLYTREK|TECENTRIQ|LORBRENA|METHOTREXATE|PEMETREXED| PEMETREXED DISODIUM|ATEZOLIZUMAB|ABRAXANE|PEMBROLIZUMAB|MVASI|DOXORUBICIN HYDROCHLORIDE|PARAPLATIN|LORLATINIB|RETEVMO|RAMUCIRUMAB|MECHLORETHAMINE HYDROCHLORIDE|GEFITINIB|AFATINIB|CAPMATINIB HYDROCHLORIDE|DURVALUMAB|NIVOLUMAB| CRIZOTINIB|KEYTRUDA|TABRECTA|ZYKADIA|AFATINIB DIMALEATE|NECITUMUMAB|ALECTINIB| ERLOTINIB ERRFI1|CRKL|STK10|STK11|MOCS3|B2M|CD226|LIFR|TAT|SMG1|PIK3CA|CAMK4| PIK3CG|RNF114|ZC3H15|WEE1|SCLY|IGF1R|XPC|MUC16|CYP3A4|CYP3A5|GATA2|CNDP1|CNDP2| DUSP28|CCND1|TUBB4B|DDX43|PTEN|STK38L|CDKN1B|CDKN1A|ALDH3A1|HSP90B1|EDNRB| SNORD68|RP11-162A12.2|SOX10|RP11-321M21.3|JAK2|NF2|NF1|TTL|CX3CL1|ZNF516| JAK1|DSEL|DST|NCAPD3|PPARD|VEGFA|VEGFC|VEGFB|GART|RTTN|FSIP2|BNIP3|TNK2|RPS6KA3| RPS6KA6|CBL|FLNC|HES6|TLE3|RET|PRKDC|SMO|XIAP|EML4|FH|GLS|TSC2|HTRA1|TSC1|CIITA| SPARC|S100A8|ITGAV|GGH|IL17F|GSG2|RXRA|LDLR|LTK|UBE2V1|IL2RA|ALK|SLCO1B3|TP73| PBRM1|BRCA1|MYC|MAP2K5|ERBB3|DHFR|EGFR|BAX|CTNNB1|OR6B2|OR6B3|HLA-DRA|DPYD| CDK12|TWIST2|RNPEPL1|CXCR2|HNF1A|PGP|FOLH1|NOS3|AC110619.2|RP11-169F17.1|HGF| TRIP11|TRIP13|MTERFD2|EIF4E2|E2F1|BTRC|CYB5A|PPP3CB|LIF|KIAA1549|STYK1|ARAF| FCGR1A|EGF|PIK3C3|TUBA3E|TUBA3D|FGFR3|NDUFA10|ZNF236|CDH1|FGFR1|CDH7|CDKN2A| RAMP1|SCD5|FAM69C|KLK3|TPR|SNAI1|C1QC|C1QB|CSF2|NETO1|APC|ADNP|CCL4|KCNG1|COPS5| TGM2|TRIB1|SLC29A1|HSPB2|ATP7A|ATP7B|RP11-861L17.3|CYP2D6|NEK1|STK25|EMP1| CEACAM5|TRAF3IP1|TNKS2|PRR21|IGF2|FARP2|IGFBP5|TUBB2A|TUBB2B|STK32A|TUBB|TOP1| CLK1|CLK3|GAS6|EPHX1|PDGFRA|IC4|PI4KA|TPM3|ATM|CYP4B1|TIGD1|ATR|RP11-41O4.1| FANCD2|FLT3|AREG|CYP1B1|POLD1|MYH9|TMEM200A|CDKL2|FANCG|FANCC|AC104809.3|FASN| FBXW7|ACVR2A|ANO7|RALG”S|MT'R11|NUDT10|MTOR|GTSCR1|LINC00908|EP300|MAPK8| FAM65C|KIF1A|PIP5K1B|C”K2|S'AMF1|POLA1|TGFB3|PTK7|ERCC2|PRDX4|ARMS2|MSH6|MCL1| HIP1|BAZ2B|CEBPB|”EBPA'AXIN1|RGL1|MAP2|RB1|ETS2|ABCB1|AURKA|SULT2A1|PRKCH| TOP1MT|FCGR2B|FCGR2C|FCGR2A|SKILL|TNFRSF11B|TUBA4A|BAI1|NOTCH1|NOTCH3|TYMS|GSTP1| MYEOV2|MMP9|BCL2|MMP2|FES|TSSK1B|PMAIP1|AGE|FBXO15|GPR35|PPP1R7|LDHA|PARD6B| CBFA2T3|JAK3|ERBB2|FLCN|CIP2A|BRCA2|SHC1|ERBB4|TRIM24|IDH1|WNK1|CLCN6|PASK|PER2| PER1|RPL13|BRD4|PATZ1|FGFR4|FGFR2|ROR2|SHMT1|LRRFIP1|ATIC|RBM44|ZFYVE9|SYK| CXCL8|KIT|SPEN|GALR1|FOXA2|FKBP1A|HRAS|MED24|XRCC1|XRCC3|XRCC4|SEPT3|SEPT2| UBE2F-SCLY|FOS|SLC9A8|ROS1|RRM2B|NFE2|ANXA11|CHEK1|CTLA4|TUBB8|SULT1C4|TUBB6| TSHZ1|TUBB3|TUBB1|ASB1|PTPN1|ESPNL|TP53|GNAQ|ESR1|DDR2|CD274|UBE2F|EIF4EBP1| STK38|TIE1|RP11-17M16.1|AC062017.1|NTRK1|SNED1|BCAS4|NTRK3|CHST3|KCNA1|STAT6| ATXN1L|ANKMY1|CBLN2|ICAM3|AC093802.1|MSH2|DPM1”SRC|'SH3|TSSK2|ACACB|MAST2| SALL3|ATRX|LYN|CDA|SGK1|VHL|GSTM1|GNAS|AGXT|CAPN10|IL2|C2orf54|ZNF407|C18orf63| TMPO|MAP4K5|CAMK1G|CDH19|BIRC5|FCGR3A|FCGR3B|SLCO1B1|MAP2K2|FOXO1|AC016757.3| MAP2K7|MAP2K6|MAP4|MAP2K4|ALDH1A1|MLLT4|SMARCA4|NIPBL|HNRNPA2B1|NRG1|MBP|ALOX5| MALT1|SMARCB1|COL18A1|HSPB8|MCM4|ILKAP|CYP2A6|OTOS|CASP3|INSR|BCL9|MAPT|ABCC4| C1R|ABCC6|AC079612.1|ABCC3|FAM132B|AKT1|AKT2|AKT3|ATP9B|NIM1|BDNF|S100A12|MS4A2| ENOSF1|CCDC102B|GRK5|YWHAH|NTRK2|GRK1|SMAD4|ARHGAP29|MYB|BLNK|DDIT3|P4HB| SERPINB1|PMS2|SMIM21|UBR5|KRAS|SOCS6|ROBO2|GSK3A|POLE|ZADH2|GSK3B|NR1I2| TMEM189-UBE2V1|TLR9|KDM5A|SLC19A1|STMN1|HLA-A|DCLK1|FMN2|DCP1B|ARID4B|MAPKAPK3| MAPKAPK2|CHD1|TUBA1C|TUBA1B|TUBA1A|THBS2|AFP|ETV5|TMEM189|RP11-723G8.2|TOP2A| ACVRL1|KLHL30|SLC10A2|STAG2|MAP2K3|PKHD1|ITPR1|FGF2|PBK|ETV1|IL1RN|ABCG2|TLK1| ETV4|ETV6|MPL|G”PD|N'P153|UNC79|MATK|MAP2K1|SACS|ABL1|ABL2|CACNB2|MAPK3|MAPK1| KDM6A|PDCD4|DUSP6|PDCD1|MTHFR|HDLBP|FER|UNC13C|ADA|PRB2|CRNKL1|KRT20|MET|KDR| WWOX|TIMM21|CIC|SPG7|BCL11A|LATS2|LATS1|RRM2|RRM1|DSCAM|UBASH3B|TXK|PTPRF|CMPK1| TAL2|NRAS|GCLC|MLH1|DOK6|AQP12B|AQP12A|NAT2|C1QA|DAPK1|ULK2|NR3C1|RP11-94B19.4| GNA11|PLAU|GPC1|TMX3|JARID2|NPM1|ABCC10|TAB2|CHD9|GRIN2A|BRAF|CHD8|SPATA2|BAP1| ATF1 LUSC CYRAMZA|IRESSA|AVASTIN|TARCEVA|OSIMERTINIB MESYLATE|AFINITOR| ALECTINIB HYDROCHLORIDE|PORTRAZZA|SELPERCATINIB|VINORELBINE|TOPOTECAN HYDROCHLORIDE|PORFIMER|METHOTREXATE SODIUM|ENTRECTINIB|IPILIMUMAB|CERITINIB| IMFINZI|GILOTRIF|TAFINLAR|ALIMTA|ERLOTINIB HYDROCHLORIDE|XALKORI|EVEROLIMUS| ALECENSA|GEMCITABINE|MEKINIST|PACLITAXEL|TAXOTERE|TRAMETINIB|CARBOPLATIN| PARAPLAT|AFINITOR DISPERZ|DOCETAXEL|DABRAFENIB MESYLATE|OPDIVO|TAGRISSO| BEVACIZUMAB|PACLITAXEL ALBUMIN-STABILIZED NANOPARTICLE FORMULATION|YERVOY| BRIGATINIB|GEMZAR|MUSTARGEN|TREXALL|VIZIMPRO|VINORELBINE TARTRATE|ALUNBRIG| GEMCITABINE HYDROCHLORIDE|TOPOTECAN|DACOMITINIB|ROZLYTREK|TECENTRIQ|LORBRENA| METHOTREXATE|PEMETREXED|PEMETREXED DISODIUM|ATEZOLIZUMAB|ABRAXANE|PEMBROLIZUMAB| MVASI|DOXORUBICIN HYDROCHLORIDE|PARAPLATIN|LORLATINIB|RETEVMO|RAMUCIRUMAB| MECHLORETHAMINE HYDROCHLORIDE|GEFITINIB|AFATINIB|CAPMATINIB HYDROCHLORIDE| DURVALUMAB|NIVOLUMAB|CRIZOTINIB|KEYTRUDA|TABRECTA|ZYKADIA|AFATINIB DIMALEATE| NECITUMUMAB|ALECTINIB|ERLOTINIB ERRFI1|CRKL|STK10|STK11|MOCS3|B2M|CD226|LIFR| TAT|SMG1|PIK3CA|CAMK4|PIK3CG|RNF114|ZC3H15|WEE1|SCLY|IG”1R|X’C|MUC16|CYP3A4| CYP3A5|GATA2|CNDP1|CNDP2|DUSP28|CCND1|TUBB4B|DDX43|PTEN|STK38L|CDKN1B|CDKN1A| ALDH3A1|HSP90B1|EDNRB|SNORD68|RP11-162A12.2|SOX10|RP11-321M21.3|JAK2|NF2|NF1| TTL|CX3CL1|ZNF516|JAK1|DSEL|DST|NCAPD3|PPARD|VEGFA|VEGFC|VEGFB|GART|RTTN|FSIP2| BNIP3|TNK2|RPS6KA3|RPS6KA6|CBL|FLNC|HES6|TLE3|RET|PRKDC|SMO|XIAP|EML4|FH|GLS| TSC2|HTRA1|TSC1|CIITA|SPARC|S100A8|ITGAV|GGH|IL17F|GSG2|RXRA|LDLR|LTK|UBE2V1| IL2RA|ALK|SLCO1B3|TP73|PBRM1|BRCA1|MYC|MAP2K5|ERBB3|DHFR|EGFR|BAX|CTNNB1|OR6B2| OR6B3|HLA-DRA|DPYD|CDK12|TWIST2|RNPEPL1|CXCR2|HNF1A|PGP|FOLH1|NOS3|AC110619.2| RP11-169F17.1|HGF|TRIP11|TRIP13|MTERFD2|EIF4E2|E2F1|BTRC|CYB5A|PPP3CB|LIF| KIAA1549|STYK1|ARAF|FCGR1A|EGF|PIK3C3|TUBA3E|TUBA3D|FGFR3|NDUFA10|ZNF236|CDH1| FGFR1|CDH7|CDKN2A|RAMP1|SCD5|FAM69C|KLK3|TPR|SNAI1|C1QC|C1QB|CSF2|NETO1|APC| ADNP|CCL4|KCNG1|COPS5|TGM2|TRIB1|SLC29A1|HSPB2|ATP7A|ATP7B|RP11-861L17.3| CYP2D6|NEK1|STK25|EMP1|CEACAM5|TRAF3IP1|TNKS2|PRR21|IGF2|FARP2|IGFBP5|TUBB2A| TUBB2B|STK32A|TUBB|TOP1|CLK1|CLK3|GAS6|EPHX1|PDGFRA|IC4|PI4KA|TPM3|ATM|CYP4B1| TIGD1|ATR|RP11-41O4.1|FANCD2|FLT3|AREG|CYP1B1|POLD1|MYH9|TMEM200A|CDKL2|FANCG| FANCC|AC104809.3|FASN|FBXW7|ACVR2A|ANO7|RALGDS|MTMR11|NUDT10|MTOR|GTSCR1| LINC00908|EP300|MAPK8|FAM65C|KIF1A|PIP5K1B|CDK2|SLAMF1|POLA1|TGFB3|PTK7|ERCC2| PRDX4|ARMS2|MSH6|MCL1|HIP1|BAZ2B|CEBPB|CEBPA|AXIN1|RGL1|MAP2|RB1|ETS2|ABCB1| AURKA|SULT2A1|PRKCH|TOP1MT|FCGR2B|FCGR2C|FCGR2A|SKILL|TNFRSF11B|TUBA4A|BAI1| NOTCH1|NOTCH3|TYMS|GSTP1|MYEOV2|MMP9|BCL2|MMP2|FES|TSSK1B|PMAIP1|AGE|FBXO15| GPR35|PPP1R7|LDHA|PARD6B|CBFA2T3|JAK3|ERBB2|FLCN|CIP2A|BRCA2|SHC1|ERBB4|TRIM24| IDH1|WNK1|CLCN6|PASK|PER2|PER1|RPL13|BRD4|PATZ1|FGFR4|FGFR2|ROR2|SHMT1|LRRFIP1| ATIC|RBM44|ZFYVE9|SYK|CXCL8|KIT|SPEN|GALR1|FOXA2|FKBP1A|HRAS|MED24|XRCC1|XRCC3| XRCC4|SEPT3|SEPT2|UBE2F-SCLY|FOS|SLC9A8|ROS1|RRM2B|NFE2|ANXA11|CHEK1|CTLA4| TUBB8|SULT1C4|TUBB6|TSHZ1|TUBB3|TUBB1|ASB1|PTPN1|ESPNL|TP53|GNAQ|ESR1|DDR2| CD274|UBE2F|EIF4EBP1|STK38|TIE1|RP11-17M16.1|AC062017.1|NTRK1|SNED1|BCAS4| NTRK3|CHST3|KCNA1|STAT6|ATXN1L|ANKMY1|CBLN2|ICAM3|AC093802.1|MSH2|DPM1|SRC|MSH3| TSSK2|ACACB|MAST2|SALL3|ATRX|LYN|CDA|SGK1|VHL|GSTM1|GNAS|AGXT|CAPN10|IL2| C2orf54|ZNF407|C18orf63|TMPO|MAP4K5|CAMK1G|CDH19|BIRC5|FCGR3A|FCGR3B|SLCO1B1| MAP2K2|FOXO1|AC016757.3|MAP2K7|MAP2K6|MAP4|MAP2K4|ALDH1A1|MLLT4|SMARCA4|NIPBL| HNRNPA2B1|NRG1|MBP|ALOX5|MALT1|SMARCB1|COL18A1|HSPB8|MCM4|ILKAP|CYP2A6|OTOS| CASP3|INSR|BCL9|MAPT|ABCC4|C1R|ABCC6|AC079612.1|ABCC3|FAM132B|AKT1|AKT2|AKT3| ATP9B|NIM1|BDNF|S100A12|MS4A2|ENOSF1|CCDC102B|GRK5|YWHAH|NTRK2|GRK1|SMAD4| ARHGAP29|MYB|BLNK|DDIT3|P4HB|SERPINB1|PMS2|SMIM21|UBR5|KRAS|SOCS6|ROBO2|GSK3A| POLE|ZADH2|GSK3B|NR1I2|TMEM189-UBE2V1|TLR9|KDM5A|SLC19A1|STMN1|HLA-A|DCLK1|FMN2| DCP1B|ARID4B|MAPKAPK3|MAPKAPK2|CHD1|TUBA1C|TUBA1B|TUBA1A|THBS2|AFP|ETV5|TMEM189| RP11-723G8.2|TOP2A|ACVRL1|KLHL30|SLC10A2|STAG2|MAP2K3|PKHD1|ITPR1|FGF2|PBK|ETV1| IL1RN|ABCG2|TLK1|ETV4|ETV6|MPL|G6PD|NUP153|UNC79|MATK|MAP2K1|SACS|ABL1|ABL2| CACNB2|MAPK3|MAPK1|KDM6A|PDCD4|DUSP6|PDCD1|MTHFR|HDLBP|FER|UNC13C|ADA|PRB2| CRNKL1|KRT20|MET|KDR|WWOX|TIMM21|CIC|SPG7|BCL11A|LATS2|LATS1|RRM2|RRM1|DSCAM| UBASH3B|TXK|PTPRF|CMPK1|TAL2|NRAS|GCLC|MLH1|DOK6|AQP12B|AQP12A|NAT2|C1QA|DAPK1| ULK2|NR3C1|RP11-94B19.4|GNA11|PLAU|GPC1|TMX3|JARID2|NPM1|ABCC10|TAB2|CHD9| GRIN2A|BRAF|CHD8|SPATA2|BAP1|ATF1 PRAD PROVENGE|RADIUM 223 DICHLORIDE|RUBRACA|ELIGARD|XTANDI|FIRMAGON| FLUTAMIDE|ENZALUTAMIDE|NILUTAMIDE|LYNPARZA|ERLEADA|DEGARELIX ACETATE| CABAZITAXEL|MITOXANTRONE|ABIRATERONE|NILANDRON|JEVTANA|LEUPROLIDE ACETATE| BICALUTAMIDE|LUPRON DEPOT|RUCAPARIB CAMSYLATE|GOSERELIN|GOSERELIN ACETATE| XOFIGO|MITOXANTRONE HYDROCHLORIDE|LUPRON|DEGARELIX|MELPHALAN|ESTRAMUSTINE| TAXOTERE|ZOLADEX|OLAPARIB|DAROLUTAMIDE|ESTRAMUSTINE PHOSPHATE SODIUM|DOCETAXEL| NUBEQA|CASODEX|ABIRATERONE ACETATE|APALUTAMIDE|SIPULEUCEL-T|ESTRAMUSTINE PHOSPHATE|ZYTIGA GSR|STK11|PEBP1|FST|TUBA3E|TUBA3D|CDH1|PIK3CA|DNMT1|PZP|IRAK2| CBLC|BRCA2|TH|IL2RA|IFNG|IDH2|IDH1|AQP9|KLK3|RPL13|BRD4|MUC16|CYP3A4|CYP3A5| ROR2|KRAS|OPLAH|TUBB4B|AHR|PTEN|VIP|ATR|TOP2A|VDR|MAP1A|HRAS|ABL1|MGMT|XRCC1| ARID4B|ATP7A|XRCC4|LHCGR|NFE2L2|NR1I2|ASXL1|TUBA1B|TUBA1A|TUBA1C|PPARGC1A|TUBB6| CDKN1A|CHEK2|CHEK1|IGF2|”TLA4’TUBB8|SULT1C4|STAG2|PARP1|PARP2|TUBB3|PPARD| CHST3|TUBB1|TUBB2A|TUBB”B|AB’G2|TP53|CSNK1A1|ESR1|MPO|PALB2|TUBB|PCDH15|GAS6| EPHX1|HDAC2|HDAC4|TLE3|EPHA4|CYP4B1|SMAD3|GNRHR|FH|SLC10A2|PRAME|CACNB2|CYP2B6| CFLAR|FANCC|FANCA|MSH2|FBXW7|GNRHR2|MTHFR|NAT2|HDLBP|DLL1|ESR2|ADK|SLCO1B1|ACPP| RPSA|CYP17A1|GSTM1|MET|GNRH1|RIPK4|BRCA1|TOP2B|RRM1|SPG7|KALRN|BIRC5|PRDX4|BAX| MAP2|VHL|SLCO1B3|MCL1|MAP4|UGT1A5|XRCC3|MYOD1|ATM|MLLT6|SNORD68|ABCB1|PGP”LDHA 'FOLH1|TRAF2|TET2|KMT2D|PRL|PLAT|AR|CYP2A6|TUBA4A|ERBB2|HEXB|NOTCH1|PARP3| CDK12|GSTP1|PTPRC|MAPT|ABCC6|MMP1|ABCC1|BCL2 SKCM SONIDEGIB|FLUOROURACIL-TOPICAL|COBIMETINIB|VINBLASTINE SULFATE|ERIVEDGE| VISMODEGIB|PACLITAXEL|VORINOSTAT|TRAMETINIB|ODOMZO|ALDARA|COBIMETINIB FUMARATE| DACARBAZINE|DABRAFENIB MESYLATE|VEMURAFENIB|1218778-77-8|ERISMODEGIB| DABRAFENIB|1187431-43-1|5-FU|VINBLASTINE|EFUDEX|ALITRETINOIN|IMIQUIMOD| AVELUMAB PDE4DIP|LIF|CRKL|STK11|PCSK6|HDAC3|PTHLH|AKT1|ARAF|TNRC6B|HIPK4|BDNF| PI4KB|RARA|RARB|TUBA3E|TUBA3D|RARG|BRCA1|PIK3CA|LRP1B|SYNE1|PIK3CG|SMAD4|PHC1| EPHA6|PRKG1|MYC|MYB|ERBB2|DDIT3|TPO|IDH2|KLK3|MGMT|CYP3A4|CYP3A5|FGFR2|FGFR3| TBXA2R|KRAS|SMO|SYK|CXCL8|DDX43|ROBO2|STAG2|CSF2|KIT|PTEN|TLR7|TUBB4B|IC11|IC10| TLR8|TUBE1|BRCA2|STMN1|VKORC1|CDKN1A|DCLK2|ABL1|CRTC3|EDNRB|NEK11|NPRL2|MYCN| TCF3|NR1I2|TUBA1C|TUBA1B|TUBA1A|JUN|MKL1|EPHA7|TIAM1|CTNNB1|JAK2|NFE2|NF1|TUBD1| TRPM6|MYD88|TOP2A|TRH|EPHB1|TUBB8|FOS|TUBB6|STAG3|TUBB3|HDAC9|TUBB1|BLM|MAP2K3| RHOH|HRAS|ATXN1L|PLAT|TUBB2A|ING1|TUBB2B|ETV1|TP53|GNAQ|RPS6KA6|CD274|G6PD| NYNRIN|TUBB|EIF4EBP1|IC1|PDGFRB|PDGFRA|HDAC2|HDAC5|HDAC4|HDAC7|HDAC6|TLE3|HDAC8| ATM|SRGAP3|TIGD1|AKAP12|SACS|FH|WDHD1|RAF1|NCOA3|CYP1B1|BTG1|CGA|CSF1R|NR2F2| SPARC|FASN|NBN|TBX18|PDCD4|DUSP6|ERCC2|FANCI|FBXW7|SRC|MTHFR|NAT2|RXRG|RXRA| RXRB|ETV5|ADK|ETV4|IL2RA|ITK|MAP2K2|IFNA1|MET|CDK2|EZH2|FGFR1|CIC|POLA2|MAP4| MAP2K5|ACVR2A|KALRN|BIRC5|LIMK1|EGFR|BAX|PTCH1|MAP2|VHL|MAP2K1|CDKN2A|MAP2K7| MAP2K6|RRM1|MAP2K4|MITF|DSCAM|ARID4B|TEAD4|PGD|SIK1|NRAS|CDK17|RB1|ETS2|ABCB1| AURKA|PGP|ABCB4|NRG1|C15orf2|EPM2AIP1|ULK2|PRKCH|SMARCB1|RAC1|PLAU|FCGR2A|MCM3| SUFU|TUBA4A|NPM1|ABCC10|EIF4E2|CASP3|E2F1|NOTCH1|NOTCH2|TUBG1|SOX10|GSTP1|BRAF| MAPT|GNA11|BAP1|ABCC3|MMP2|BCL2 STAD CYRAMZA|DOXORUBICIN HYDROCHLORIDE|TRIFLURIDINE AND TIPIRACIL HYDROCHLORIDE|PEMBROLIZUMAB|EVEROLIMUS|5-FU|RAMUCIRUMAB|TRASTUZUMAB|TAXOTERE| HERCEPTIN|MITOMYCIN|SUNITINIB|SUNITINIB MALATE|LANREOTIDE|DOCETAXEL| MITOMYCIN C|KEYTRUDA|LONSURF|REGORAFENIB|FLUOROURACIL INJECTION ERBB3|PDGFRB| STK11|AKT1|AKT2|AKT3|ARAF|MLH1|FCGR1A|PMS2|TAT|B2M|PIK3CA|NTRK1|PIK3CG|EDN1| FGFR1|BRCA1|FLCN|BRCA2|AURKB|IFNG|KRT20|EPHA2|BGLAP|KLK3|RPL13|BRD4|MUC16| CYP3A4|CYP3A5|FGFR2|KRAS|ARID1A|C1QC|C1QB|C1QA|KIT|NR1I2|XRN1|FKBP1A|APC|HRAS| HIF1A|ABL1|PTPRB|XRCC1|HSPB2|ARID4B|ATP7A|XRCC4|PTEN|U2AF1|RAF1|XDH|JAK2|NF2| NF1|JAK1|FLT1|TOP2A|IGF2|SULT1C4|TUBB3|TUBB1|PPARD|CHST3|TP53|GNAQ|ESR1|DDR2| CD274|MPL|HMOX1|TUBB|VWF|GAS6|EPHX1|PDGFRA|MKI67|TLE3|ATM|CYP4B1|RET|POLE| MAPK10|MAPK11|FH|IL1A|FLT4|TSC2|FLT3|TSC1|SLC10A2|TEK|CSF1R|FANCC|MSH2|KDR| FBXW7|MSH3|MTHFR|NAT2|CA9|CGB5|AGAP1|AC064874.1|MTOR|MSH6|LGALS2|SLCO1B1|ALK| GSTM1|SSTR1|MET|SSTR3|SSTR2|SSTR5|PBRM1|PDCD1|PALB2|ERBB2|MAP4|SPG7|DHFR|BIRC5| PRDX4|EGFR|BAX|FCGR3B|CTNNB1|VHL|SLCO1B3|MCL1|RRM1|FCGR3A|XRCC3|HLA-DRA|DPYD| MAP2|NRAS|RB1|SNORD68|FRK|ABCB1|PGR|PGP|ABCB4|FOLH1|ANXA1|PDGFA|PDGFB|PDGFC| PDGFD|FCGR2B|FCGR2C|FCGR2A|CYP2A6|HGF|EPCAM|BAP1|CYP7B1|NOTCH1|TYMS|GSTP1|BRAF| MAPT|GNA11|C1S|C1R|ABCC6|KRT18|BCL2 UCECHYDROXYPROGESTERONE CAPROATE|DACTINOMYCIN|VINBLASTINE|METHOTREXATE SODIUM|LENVIMA|PEMBROLIZUMAB|MEGESTROL ACETATE|KEYTRUDA|METHOTREXATE|VINBLASTINE SULFATE|LENVATINIB MESYLATE SALL3|PCSK6|FBXO15|DUSP28|B2M|CD226|RB1|BDNF| S100A12|GPR35|C18orf63|TAT|PPP1R7|ASB1|TUBA3E|TUBA3D|CCDC102B|PIK3CB|NTRK1| ANKMY1|NDUFA10|ZNF236|FAM132B|JAK2|MYC|SCLY|CDH7|DDIT3|IDH2|ALOX5|RAMP1|CLCN6| PASK|FAM69C|BGLAP|KLK3|COIL|PMS2|SMIM21|CNDP1|CNDP2|KRAS|ATIC|RBM44|CCND1| ZNF516|CSF1|CSF2|KIT|PTEN|GALR1|GRB2|TUBB4B|NETO1|SULT2A1|TUBE1|SLC19A1|CDKN1B| CDKN1A|ANO7|SNED1|CRTC3|ZADH2|AC110619.2|ARID4B|RP11-162A12.2|UBE2F-SCLY| RP11-861L17.3|TUBB8|NR1I2|TUBA1C|TUBA1B|TUBA1A|CEL|JUN|RP11-94B19.4|STK25| RP11-321M21.3|TNFRSF8|NF1|JAK1|TUBD1|TRAF3IP1|FLT1|RP11-723G8.2|PRR21|TOP2A| DSEL|TRH|FARP2|KLHL30|TUBB6|TSHZ1|TUBB3|ESPNL|BLM|TUBB1|TUBB2A|TUBB2B|IL1RN| TP53|CD274|ATP9B|TUBB|UBE2F|TIMM21|OTOS|HES6|HDAC4|CBLN2|AC062017.1|RNPEPL1|CD5| POLE|RP11-41O4.1|FH|FLT4|GLS|FLT3|KCNA1|CGA|AFP|TPM3|NR2F2|CDH19|ICAM3|S100A8| TFPI|KIF1A|MSH2|FANCI|PDCD1|TBXA2R|NTF3|MTHFR|HDLBP|RTTN|BAX|GTSCR1|ADA| LINC00908|TNFRSF1B|IL2RA|ALK|GSTM1|AGXT|IL15|CAPN10|PGR|IL2|C2orf54|ZNF407| SLAMF1|RP11-17M16.1|DHFR|BIRC5|MSH3|EGFR|MSH6|FCGR3B|SLCO1B1|AC016757.3| AC104809.3|OR6B2|OR6B3|HLA-DRA|AC079612.1|TWIST2|NRAS|PTHLH|MLH1|ABCB1|DOK6| AQP12B|AQP12A|ABCB4|NRG1|SEPT2|MBP|LRRFIP1|FOLH1|NR3C1|HSPB2|MTR|RP11-169F17.1| GPC1|SOCS6|ILKAP|TMX3|HGF|TUBA4A|CYB5A|MTERFD2|E2F1|PER2|NOTCH1|TUBG1|TYMS| AC093802.1|KDR|MYEOV2|BRAF|ABCC4|MMP2 UCS HYDROXYPROGESTERONE CAPROATE|DACTINOMYCIN|VINBLASTINE|METHOTREXATE SODIUM|LENVIMA|PEMBROLIZUMAB|MEGESTROL ACETATE|KEYTRUDA|METHOTREXATE|VINBLASTINE SULFATE|LENVATINIB MESYLATE SALL3|PCSK6|FBXO15|DUSP28|B2M|CD226|RB1|BDNF| S100A12|GPR35|C18orf63|TAT|PPP1R7|ASB1|TUBA3E|TUBA3D|CCDC102B|PIK3CB|NTRK1| ANKMY1|NDUFA10|ZNF236|FAM132B|JAK2|MYC|SCLY|CDH7|DDIT”|IDH’|ALOX5|RAMP1|CLCN6| PASK|FAM69C|BGLAP|KLK3|COIL|PMS2|SMIM21|CNDP1|CNDP2|KRAS|ATIC|RBM44|CCND1| ZNF516|CSF1|CSF2|KIT|PTEN|GALR1|GRB2|TUBB4B|NETO1|SULT2A1|TUBE1|SLC19A1|CDKN1B| CDKN1A|ANO7|SNED1|CRTC3|ZADH2|AC110619.2|ARID4B|RP11-162A12.2|UBE2F-SCLY| RP11-861L17.3|TUBB8|NR1I2|TUBA1C|TUBA1B|TUBA1A|CEL|JUN|RP11-94B19.4|STK25| RP11-321M21.3|TNFRSF8|NF1|JAK1|TUBD1|TRAF3IP1|FLT1|RP11-723G8.2|PRR21|TOP2A| DSEL|TRH|FARP2|KLHL30|TUBB6|TSHZ1|TUBB3|ESPNL|BLM|TUBB1|TUBB2A|TUBB2B|IL1RN| TP53|CD274|ATP9B|TUBB|UBE2F|TIMM21|OTOS|HES6|HDAC4|CBLN2|AC062017.1|RNPEPL1|CD5| POLE|RP11-41O4.1|FH|FLT4|GLS|FLT3|KCNA1|CGA|AFP|TPM3|NR2F2|CDH19|ICAM3|S100A8| TFPI|KIF1A|MSH2|FANCI|PDCD1|TBXA2R|NTF3|MTHFR|HDLBP|RTTN|BAX|GTSCR1|ADA| LINC00908|TNFRSF1B|IL2RA|ALK|GSTM1|AGXT|IL15|CAPN10|PGR|IL2|C2orf54|ZNF407| SLAMF1|RP11-17M16.1|DHFR|BIRC5|MSH3|EGFR|MSH6|FCGR3B|SLCO1B1|AC016757.3| AC104809.3|OR6B2|OR6B3|HLA-DRA|AC079612.1|TWIST2|NRAS|PTHLH|MLH1|ABCB1|DOK6| AQP12B|AQP12A|ABCB4|NRG1|SEPT2|MBP|LRRFIP1|FOLH1|NR3C1|HSPB2|MTR|RP11-169F17.1| GPC1|SOCS6|ILKAP|TMX3|HGF|TUBA4A|CYB5A|MTERFD2|E2F1|PER2|NOTCH1|TUBG1|TYMS| AC093802.1|KDR|MYEOV2|BRAF|ABCC4|MMP2

[0064] 1-24. Calculating TC Score A therapeutic candidate (TC) score was assigned to each drug for 17 cancer types. The TC score was calculated by taking the root mean square of the network propagation values ​​of the drug's target and biomarker genes. The calculation formula is as follows:

number

[0065] 1-25. Drug cytotoxicity prediction performance To obtain drug response data for human cancer cell lines, the inventors downloaded the "secondary-screen-dose-response-curve-parameters" from the PRISM database. This data includes 1,448 compounds screened from 499 cell lines. Of the 17 cancer types, 15 cancer types, i.e., cancer cell lines, that have secondary screen data in PRISM were used. The cytotoxic effect of a drug in each cancer type was determined by the drug's median IC20 in the cell line belonging to that cancer. 50 was calculated by value.

[0066] The performance of the network is evaluated by the TC score of the drug for each cancer and the median IC of the drug for that cancer. 50 The correlation coefficient between the negative logarithm of the values ​​was measured using the Spearman correlation coefficient. Drug synonyms between PRSIM and PanDrugs were mapped using PubChem IDs through the Pubchempy Python module.

[0067] 1-26. Performance of drug reversal gene expression (RGE) prediction in COADREAD To evaluate the efficacy of drugs by altering gene expression in cancerous conditions, we used the L1000 dataset from Cmap [9]. In the L1000 dataset, we obtained expression profiles for 6,056 compounds with PubChem IDs and 978 genes in 15 cancer cell lines (CL34, HCT116, HELA, HT115, HT29, LOVO, MDST8, NCIH508, NCIH716, RKO, SNU1040, SNUC5, SW480, SW620, and SW948) predominantly distributed in the colon. We first calculated the RGE effect between COADREAD differential gene expression (obtained through CREEDS

[10] , signature ID: dz552) and the compound transcriptional profile. Here, the RGE effect was calculated using Zhang's connection score

[11] , which is the absolute value of the contrastive and normalized ranked connection score. Because drugs have many transcriptional profiles for each cell line depending on the treatment dose and time, we calculated the maximum RGE effect for each cell line and used the median RGE effect of drugs commonly present in PanDrugs and Cmap as a representative value. Finally, the performance of the network was measured using the Spearman correlation coefficient between the TC score of a drug and the RGE effect of drugs commonly present in PanDrugs and Cmap.

[0068] 1-27. Selection of candidates for drug use change From the 1,702 approved drugs in Pandrugs

[12] , the inventors selected drugs with a rank percentage of TC score of 0.9 or higher, thereby selecting 333 approved drugs with new therapeutic guidelines for 17 cancer types.

[0069] For further experimental validation in LIHC, we selected ixazomib citrate, which had the highest TC score, from among the repurposing candidates predicted by the co-expression network.

[0070] 1-28.Cell culture Human LIHC cell lines SNU398, SK-HEP-1, and Huh7 were obtained from the Korean Cell Line Bank (Seoul, Korea), and HepG2 cells were obtained from the American Type Culture Collection (Manassas, VA, USA). SNU398 and Huh7 were cultured in Roswell Park Memorial Institute (RPMI) 1640 medium (Welgene) supplemented with 10% fetal bovine serum (FBS, Merck), 100 IU / mL penicillin, and 100 μg / mL streptomycin (Welgene). SK-HEP-1 and HepG2 were maintained in Eagle's Minimum Essential Medium (EMEM, Lonza) supplemented with 10% FBS. All cells were cultured at 37°C in a 5% CO2 atmosphere.

[0071] 1-29. Cell viability analysis To assess cell viability, SNU398, SK-HEP-1, Huh7, and HepG2 cells were plated in 96-well plates at 8 × 10 3 or 1 x 10 4Cells were cultured at a density of 100 cells / well. After 24 hours of incubation, the medium was removed, and the cells were maintained in 100 μL of medium supplemented with various concentrations of ixazomib citrate (MedChemExpress). After 72 hours of incubation, cell viability was determined via a phenazine methosulfate (PMS) / 3-(4,5-dimethylthiazol-2-yl)-5-(3-carboxymethoxyphenyl)-2-(4-sulfophenyl)-2H-tetrazolium, inner salt (MTS) assay using the Cell Titer 96 Aqueous Non-Radioactive Cell Proliferation Assay kit (Promega, Madison, WI, USA) according to the manufacturer's instructions. Absorbance at 490 nm was measured using a microplate reader (Multiskan SkyHigh Microplate Spectrophotometer, Thermo Scientific). IC 50 Values ​​were calculated from dose-response curves using GraphPad Prism 9 (GraphPad Software).

[0072] 1-30. Crowd formation experiment SNU398, SK-HEP-1, Huh7, and HepG2 cells were plated in 24-well plates at 4 × 10 2 or 6 x 10 2 Cells were cultured at a density of 1000 cells / well. After 24 hours of incubation, the medium was removed, and the cells were maintained in the presence of 500 μL of medium supplemented with various concentrations of ixazomib citrate. The medium was changed every 2 days. After 7 days, colonies were fixed with 10% aldehyde and stained with 1% crystal violet.

[0073] Example 2. Creation of a co-essentiality network for anti-cancer drug discovery To construct a network for identifying therapeutic targets and drug repurposing candidates for each cancer type, we inferred co-essentiality links from the correlation of gene essentiality profiles in cancer cells (Figure 1A). The co-essentiality network was constructed using a CRISPR screening dataset derived from the dependency map (Depmap) project dataset, with 18,119 genes and 8,105,180 co-essentiality links based on gene-level essentiality scores. Co-essentiality links indicate the similarity of essentiality between two genes, as measured using the context likelihood of relatedness (CLR) algorithm. To confirm whether the co-essentiality links for essentiality information are associated with biological functions, we examined whether the co-essentiality links obtained in this study were enriched in the curated pathways of interest. Such co-essentiality links indicate a strong functional relationship between two genes and are associated with similar growth phenotypes in various cancer cell types (Figures 1B and 2). For example, in KEGG pathways, gene pairs with larger link weights were more likely to be in the same pathway (Figure 1B, black dots). This tendency was higher than expected by random analysis (Figure 1B, gray dots). Similarly, gene pairs with larger link weights were more likely to be in the same biological module in five independent datasets: CORUM protein complexes, REACTOME molecular pathways, and Gene Ontology annotations [GO: BP (biological process), MF (molecular function), and CC (cellular components)].

[0074] Example 3. Performance of co-essentiality networks to identify cancer-related pathways We discovered that co-essentiality networks delineate cancer-related pathways (CRPs) associated with cancer development and cancer characteristics. To confirm the benefits of using co-essentiality networks, we compared the co-essentiality network with three other molecular networks: the PPI network (BioGRID; 18,708 nodes; 434,527 links), the co-expression network (19,120 nodes; 12,759,793 links), and the co-methylation network (16,333 nodes; 10,193,089 links). The co-essentiality network showed a high enrichment of CRPs (Figure 1C, left panel), with an odds ratio of 4.65 compared with the PPI network (odds ratio = 0.225), the co-expression network (odds ratio = 0.133), and the co-methylation network (odds ratio = 1.50). The co-essentiality link showed a significant enrichment for CRPs (initial measurement test p-value = 3.12 × 10 -3 ), whereas other molecular links showed no significant enrichment for CRPs. The co-essentiality link had the highest relative modularity among the 15 CRPs, which was greater than the randomly expected value of 6.83 CRPs (right side of Figure 1C, Table 3).

[0075] For example, in the mTOR signaling pathway (KEGG: hsa04150), which is targeted by many anticancer drugs (Figure 1D, relative modularity for the co-essentiality network, BioGRID, co-expression network, and co-methylation network; rank percentage values ​​for each relative modularity = 0.898, 0.697, 0.177, and 0.109), the co-essentiality network showed higher modularity than other networks. To confirm whether co-essentiality links could be better discovered between CRP gene pairs, we measured the relative modularity of genes in each KEGG pathway and compared it with three other molecular networks (Table 3). Because modularity is biased by node precision, we used precision-adjusted random nodes and normalized precision-controlled modularity measures. These results suggest that co-essentiality links can detect more cancer causal relationships than other molecular links and may be useful for identifying genes that play important roles in cancer treatment.

[0076] Example 4. High modularity of cancer driver genes within co-essentiality networks To confirm whether co-essentiality links are effective in identifying therapeutic targets in cancer, we investigated the modularity of disease genes in the network. Disease gene modularity is an index that represents the compatibility of information associated with a specific disease phenotype. We used cancer driver genes that cause disease-inducing mutations as disease genes. Co-essentiality links formed more dense modules compared to other molecular links. In 17 of the 19 cancer types, the modularity m of driver genes was highest in the co-essentiality network compared to the other three molecular networks (Figure 3A). We used degree-controlled modularity measurements, which were normalized by randomization tests. For example, in the case of lung squamous cell carcinoma (LUSC), the co-essentiality network had the highest modularity for driver genes (m co-essentiality Network = 15.61, m in all molecular networks PPI-BioGRID =3.38, m co-expression Network = 3.33 and m co-methylation (Network = 1.99) (Figures 3B-E and 4). For example, when comparing the LUSC driver genes (Figure 3B) between networks (Figures 3C-E), the subnetworks of driver genes were more densely connected in the co-essentiality network. FAT1 and FGFR2 were connected to the subnetwork of LUSC driver genes in the co-essentiality network but were not connected to other molecular networks. FAT1 has co-essentiality interactions with RASA1, CUL3, and ARHGAP35. FGFR2 has co-essentiality interactions with FAT1. In fact, FAT1 is a biomarker for immune checkpoint blockade in lung cancer, and inhibition of FGFR2 is also effective in treating lung cancer. The 17 cancer types with the highest modularity in the co-essentiality network were selected for further analysis in precision medicine studies.

[0077] To verify the robustness of our results, we investigated the modularity of driver genes using the Cancer Gene Census (CGC), another driver gene resource. In this case, the modularity of driver genes in the co-essentiality network was highest in 12 of the 19 cancer types (Figure 5). We also used the clustering coefficient, another measure of modularity that places more emphasis on link connectivity. In 16 of the 19 cancer types, the clustering coefficient of driver genes in the co-essentiality network was higher than in other molecular networks (Figure 6). We also needed to determine whether the modularity of driver genes changed depending on the number of cell lines used to construct the co-essentiality network. By incorporating more cancer cell lines, the co-essentiality network improved, identifying more disease modules that could be used as potential therapeutic targets (Figure 7).

[0078] The inventors also confirmed whether the high modularity of co-essentiality links based on guilt-by-association could be utilized to identify driver genes for each cancer type through linkage with known driver genes (Figure 3F). To confirm whether co-essentiality links were more suitable than other molecular links, the co-essentiality network outperformed the other three molecular networks in identifying driver genes based on association in 15 of 17 cancer types (Figure 3G). For example, in the case of head and throat cancers, the co-essentiality network outperformed the other networks in identifying driver genes by 17%, 16%, and 33%, respectively. Similarly, the guilt-by-association co-essentiality network performed better in 13 of 17 cancer types than seven other PPI networks (Figure 8): BioPlex

[13] , GPSnet

[14] , HURI

[15] , Inbiomap

[16] , iRefIndex

[17] , Pathway Commons

[18] , and STRING

[19] . Furthermore, the co-essentiality network performed better in identifying driver genes than other co-essentiality networks (PMID: 33859415)

[20] and gene interaction networks based on synthetic-lethal relationships in a previous study (Figure 9)

[21] .

[0079] Using two additional reported methods for driver gene identification, we confirmed that co-essentiality links showed the greatest performance improvement for driver gene identification and provided a robust advantage for cancer driver gene identification. Driver gene identification performance was measured using network propagation methods such as Hotnet2

[22] (Figure 10A) and uKIN

[23] (Figure 11A). In both methods, co-essentiality links identified driver genes better than other molecular links. In 8 of the 11 cancer types used in the Hotnet2 study, the co-essentiality network using the Hotnet2 algorithm performed better in driver gene identification than the 12 networks used in the uKIN study (Figure 10B). Similarly, when using the uKIN algorithm, the co-essentiality network performed at least second in 16 of the 24 cancer types used in the uKIN study (Figure 11B). Taken together, the results that the co-essentiality network contains more relationships between cancer driver genes suggest that co-essentiality links are more relevant to cancer phenotypes than other molecular links.

[0080] Example 5. Differentiation of survival rates by patient group for driver modules in co-essentiality networks To investigate the clinical relevance of driver gene modularity in the co-essentiality network, we evaluated whether network modules derived from these driver genes could be useful for understanding cancer prognosis in relation to patient survival (Figure 3H). We performed a review of 16 of the 17 TCGA cancer types for which sufficient patient survival data was available. Compared with other molecular networks, the driver modules in the co-essentiality network were more useful for classifying patients based on overall survival (Figure 3I and Figure 12). In each network, patients were divided into two groups based on the overall expression of the driver modules, and the significance of the difference in survival between the two groups was measured using a log-rank test.

[0081] Co-essentiality links provided the best driver modules for classifying patients into groups with different survival rates in 12 of the 16 cancer types. In contrast, driver modules from other molecular networks failed to classify patients for 13 cancer types in the PPI (BioGRID) network, 6 in the co-expression network, and 10 in the co-methylation network. For example, in LUSC, the "Non-Receptor Tyrosine Kinases Signaling Pathway" (R-HSA-9006927), which showed significant proximity to driver genes in the co-essentiality network, was also significantly close to driver genes in the co-essentiality network (normalized abundance score [NES] = 4.13, FDR = 9.99 × 10). -4 17A), patients were classified through down regulation and showed longer overall survival than other patients (p-value = 3.75 x 10 -4, log-rank test; Figures 17B and 17C). In contrast, co-expression and co-methylation links represented distinct driver modules in ovarian tumor domain proteases (R-HSA-5689896) and disease-specific FGFR2 signaling (R-HSA-5655253), respectively, and failed to differentiate overall patient survival (p-value = 0.0255, co-expression; p-value = 0.314, co-methylation, Figures 14 and 15). PPIs did not show significant proximity to driver genes in any biological pathway (Figure 13).

[0082] Additionally, it was shown that driver modules based on co-essentiality links provide better information about patient survival than driver genes themselves. Using only driver genes themselves without identifying modules using co-essentiality links failed to differentiate survival outcomes in 15 of 16 cancer types (bottom panel of I in Figure 3 and Figure 16). For example, in LUSC, patient groups based on the expression levels of driver modules using co-essentiality links showed a significant difference in overall survival (C in Figure 17, p-value = 3.75 x 10 -4 In contrast, there was no significant difference in overall survival between patient groups divided according to the expression level of the driver gene itself (Figure 17D, [LUSC]; p-value = 0.231).

[0083] Example 6. Prioritizing therapeutic targets for specific cancer types through co-essentiality networks To explore the therapeutic potential of the co-essentiality network, we validated the potential of co-essentiality links in cancer repurposing. We used network propagation to examine genes (potential anti-cancer drug targets) adjacent to driver genes in the co-essentiality network and used these to repurpose approved drugs (Figure 18A). To determine whether the co-essentiality network could better prioritize anti-cancer drug targets, we validated the performance of the co-essentiality network in three ways: (a) prioritization of FDA-approved anti-cancer drug targets; (b) drug cytotoxicity (IC) of cancer cells. 50 ) prediction; (c) reversal gene expression (RGE) effect prediction of drugs in cancer cells. Finally, we performed cell viability evaluation experiments in human LIHC cell lines using ixazomib citrate, a candidate for repurposing hepatocellular carcinoma (LIHC) predicted by the co-essentiality link.

[0084] Network propagation using co-essentiality links better prioritized target genes of FDA-approved anticancer drugs compared with the other 12 networks (Figure 18B, Figures 19 and 20). Among the 17 cancer types examined, propagation values ​​using co-essentiality links showed the most positive enrichment for targets of FDA-approved anticancer drugs. In this analysis, both direct targets and biomarkers of drugs were used. The average normalized enrichment score (NES) of co-essentiality links was 4.14, 39% higher than that of the co-expression link, which had the second highest NES (Figure 18B). The high NES relative to the propagation values ​​of co-essentiality links suggests that co-essentiality networks better prioritize anticancer drug targets. For example, in skin cutaneous melanoma (SKCM), the co-essentiality network showed an NES value of 4.18, which was higher than the NES values ​​of the PPI-BioGRID, co-expression, and co-methylation networks (1.59, 3.25, and 3.08, respectively). Of the top 50 genes with the highest propagation values ​​in the co-essentiality network, 19 were identified as validated targets, a number higher than the number of genes identified in the PPI-BioGRID, co-expression, and co-methylation networks (17, 14, and 14 genes, respectively, among the top 50 genes). Five of these 19 genes (CDKN1A, ATM, CRKL, SOX10, and RAF1) were detected because they shared various connections with the SKCM driver genes used as inputs in the propagation (Figure 18C).In particular, dabrafenib, an FDA-approved drug for melanoma, is known to inhibit the RAF proto-oncogene serine / threonine-protein kinase (RAF1). In the co-essentiality network, RAF1 has a high propagation value (1.37 × 10). -4 and 47), because RAF1 is linked to five SKCM driver genes, such as CTNNB1, NRAS, BRAF, NF1, and KRAS.

[0085] We also observed that co-essentiality links can identify targets of approved drugs that are not discovered through other molecular links. For example, SOX10, a target of vemurafenib in SKCM, was prioritized at position 34 in the co-essentiality network, but at position 9,695 in the PPI network, 6,688 in the co-expression link, and 3,461 in the co-methylation network (Figure 18D). The co-essentiality network can identify more interactions between SOX10 and known driver genes in SKCM that are not discovered through other molecular links. SOX10 shares co-essentiality links with seven driver genes in SKCM (TP53, MAP2K1, BRAF, PPP6C, RAC1, RB1, and BRD7), which are not connected through other molecular networks. This suggests that co-essentiality links can help identify new therapeutic targets not addressed by other molecular links.

[0086] Example 7. Derivation of drug repurposing candidates using co-essentiality networks After confirming the ability to find targets of approved drugs, we further investigated whether co-essentiality links could be useful for network-based prediction of drug response assessed in large-scale pharmacogenomic screenings. Each drug was assigned a Therapeutic Candidate (TC) score by taking the root mean square (RMS) of the propagation values ​​of that drug's targets. Therefore, we considered the average effect of drug targets on a particular cancer type and the number of drug targets to eliminate bias.

[0087] The inventors confirmed that the TC score calculated with the co-essentiality link predicted the cytotoxicity of drugs to cancer cells with better performance than other molecular links. For each cancer type, drugs with a high TC score in the co-essentiality link showed high cytotoxicity to cancer cell lines of that cancer type. The co-essentiality link correlated well with the TC score and IC in 14 of the 15 cancer types. 50 The Spearman correlation coefficients between PPIs and co-expression and co-methylation links showed the highest correlation (Figure 21A), with the average Spearman correlation coefficient (Spearman R) increases for PPIs, co-expression, and co-methylation links being 9.8%, 8.8%, and 43%, respectively. For example, TAK-733, a MEK1 / 2 inhibitor currently in clinical trials for advanced non-hematologic malignancies and advanced metastatic melanoma, is predicted to have anticancer effects in colorectal cancer (COADREAD; Figure 21B; TC score = 5.39 x 10). -3 , rank percentage value = 98.57%), and indeed showed high cytotoxicity in the COADREAD cancer cell lines (median IC 50 =1.03×10 -3In the co-essentiality network, TAK-733 has seven targets, MAP2K1, MAP2K2, BRAF, KRAS, NRAS, PIK3CA, and GNA11 (FIG. 21C, bold circles), interconnected by nine driver genes in colorectal cancer (CTNNB1, SOX9, TP53, APC, TCF7L2, TGF1, FBXW7, PTEN, and ARID1A; FIG. 21C, gray). Similarly, the co-essentiality network had higher TC scores and IC scores in 12 of 15 cancer types compared to the seven other PPI networks. 50 The co-essentiality network also showed higher correlations than the other co-essentiality networks and genetic interaction networks in all 15 cancer types (Figure 23).

[0088] Although it may be doubtful whether drug response can be predicted by driver genes alone without network propagation, the inventors found that the co-essentiality network of TC scores was not more predictive than that predicted by driver genes alone (Figure 21A, driver only). The TC score was more predictive than driver genes for all 17 cancer types. 50 had a greater correlation with the drug targets, with a 42% increase in Spearman correlation coefficient. This was because driver genes accounted for only 4.4% of the drug targets compared to when network propagation was used, in which all genes in all networks were assigned a propagation score.

[0089] TC score and IC with emphasis on COADREAD 50Since the correlation between the two was the greatest, we further investigated whether drugs with high TC scores could alter gene expression patterns opposite to the cancer state (Figure 21D). We confirmed the validity of the TC score by measuring the reversal effect of drugs on cancer-related gene expression (Figure 21D). We measured the reversal gene expression (RGE) effect, which represents the expression difference between the cancer and drug-treatment states as points (Figure 21D). We also investigated the relationship between RGE and TC score. We found that the TC score of the co-essentiality network was positively correlated with the RGE effect (Figure 21E; Spearman R = 0.303, p-value = 1.98 x 10 -3 ). For example, a high TC score and -log 10 (I C 50 TAK-733, which showed a strong inverse expression effect in the expression pattern (Figure 21B, RGE effect = 0.123, percentage value = 91.18%), also showed a strong inverse expression effect in the expression pattern (Figure 21B, RGE effect = 0.123, percentage value = 91.18%). Furthermore, the TC score derived from the co-essentiality network showed a greater correlation with the RGE effect than other molecular networks (Figure 21F). These results suggest that the co-essentiality network is a better platform than all other molecular networks for identifying driver-related candidates for drug targets and has the potential to repurpose existing drugs.

[0090] Example 8. Experimental validation of drug repurposing using co-essentiality networks To identify the novel therapeutic potential of drugs approved for other uses, we performed in-silico drug repurposing using the TC scores derived from the co-essentiality network. Of a total of 1,702 approved drugs with known target information obtained from Pandrugs, 333 were predicted to exhibit high anticancer activity in at least one of 17 cancer types. Specifically, 19 repurposed drugs were assigned new targets not addressed by other networks (Figure 24A). For example, rosiglitazone maleate, a drug approved for endocrine disorders, may have novel therapeutic potential in COADREAD.

[0091] To confirm the effectiveness of in-silico drug repurposing using the co-essentiality network, the inventors performed experimental validation of ixazomib citrate, which was approved for acute myeloma in LIHC. Five repurposing candidates predicted by the co-essentiality network but not by other networks in LIHC were derived and used to evaluate the merits of the co-essentiality network. Among the five repurposing candidates, ixazomib citrate, which had the highest TC score in the co-essentiality network, was selected for experimental validation (Figure 24B; TC score = 8.38 × 10 in the co-essentiality network). -5 , rank percentage value = 92.4%). In the co-essentiality network, the 38 drug targets of ixazomib citrate formed 131 co-essentiality links with 31 LIHC driver genes, forming a more densely packed module than the other molecular networks (Figure 24C, Figure 25; PPI-BioGRID network: 59 links, co-expression network: 88 links, co-methylation network: 87 links).

[0092] To test the anticancer activity of ixazomib citrate in LIHC, we performed MTS experiments to examine the viability of four different cell lines: SNU398, SK-HEP-1, Huh7, and HepG2. Ixazomib citrate demonstrated potent anticancer effects in all of these LIHC cell lines (Figure 24D). Specifically, ixazomib citrate demonstrated micromolar-range cytotoxicity in these LIHC cell lines: IC in Huh7 cells was 1.25; ... 50 = 137 nM, IC in HepG2 cells 50 = 197 nM, IC in SK-HEP-1 cells 50 = 270 nM, and IC in SNU398 cells 50 =746 nM. We also determined whether ixazomib citrate induces sustained growth arrest through sustained colony formation experiments (Figure 24E, left). Ixazomib citrate substantially inhibited the growth of four LIHC cell lines, demonstrating a reduction in the number and size of colonies in the ixazomib citrate-treated group (Figure 24E, right). This suggests that network propagation using co-essentiality networks can identify repurposing candidates and provide novel therapeutic options.

[0093] [Cited documents] 1. Kanehisa M, Furumichi M, Tanabe M, Sato Y, Morishima K. KEGG: new perspectives on genomes, pathways, diseases and drugs. Nucleic Acids Res [Internet]. 2017 [cited 2022 Aug 4];45:D353-61. Available from: https: / / academic.oup.com / nar / article / 45 / D1 / D353 / 2605697 2. Jassal B, Matthews L, Viteri G, Gong C, Lorente P, Fabregat A, et al. The reactome pathway knowledgebase. Nucleic Acids Res [Internet]. 2019 [cited 2022 Aug 4];48:D498-503. Available from: https: / / academic.oup.com / nar / article / 48 / D1 / D498 / 5613674 3. Ashburner M, Ball CA, Blake JA, Botstein D, Butler H, Cherry JM, et al. Gene Ontology: tool for the unification of biology. Nat Genet 2000 251 [Internet]. 2000 [cited 2022 Aug 4];25:25-9. Available from: https: / / www.nature.com / articles / ng0500_25 4. Liberzon A, Subramanian A, Pinchback R, Thorvaldsdottir H, Tamayo P, Mesirov JP. Molecular signatures database (MSigDB)3.0. Bioinformatics. 2011;27:1739-40. 5. Ruepp A, Waegele B, Lechner M, Brauner B, Dunger-Kaltenbach I, Fobo G, et al. CORUM: the comprehensive resource of mammalian protein complexes-2009. Nucleic Acids Res [Internet]. 2010 [cited 2018 Dec 14];38:D497-501. Available from: http: / / www.ncbi.nlm.nih.gov / pubmed / 19884131 6. Lee I, Blom UM, Wang PI, Shim JE, Marcotte EM. Prioritizing candidate disease genes by network-based boosting of genome-wide association data. Genome Res [Internet]. 2011 [cited 2022 Aug 8];21:1109-21. Available from: https: / / genome.cshlp.org / content / 21 / 7 / 1109.full 7. Guney E, Menche J, Vidal M, Barabasi AL. Network-based in silico drug efficacy screening. Nat Commun 2016 71 [Internet]. 2016 [cited 2022 Jun 30];7:1-13. Available from: https: / / www.nature.com / articles / ncomms10331 8. McDonald ER, de Weck A, Schlabach MR, Billy E, Mavrakis KJ, Hoffman GR, et al. Project DRIVE: A Compendium of Cancer Dependencies and Synthetic Lethal Relationships Uncovered by Large-Scale, Deep RNAi Screening. Cell [Internet]. 2017 [cited 2019 Apr 12];170:577-592.e10. Available from: https: / / www.sciencedirect.com / science / article / pii / S0092867417308127 9. Lamb J, Crawford ED, Peck D, Modell JW, Blat IC, Wrobel MJ, et al. The connectivity map: Using gene-expression signatures to connect small molecules, genes, and disease. Science (80- )[Internet]. 2006 [cited 2022 Aug 5];313:1929-35. Available from: https: / / www.science.org / doi / 10.1126 / science.1132939 10. Wang Z, Monteiro CD, Jagodnik KM, Fernandez NF, Gundersen GW, Rouillard AD, et al. Extraction and analysis of signatures from the Gene Expression Omnibus by the crowd. Nat Commun 2016 71 [Internet]. 2016 [cited 2022 Aug 5];7:1-11. Available from: https: / / www.nature.com / articles / ncomms12846 11. Zhang S-D, Gant TW. A simple and robust method for connecting small-molecule drugs using gene-expression signatures. BMC Bioinformatics [Internet]. 2008 [cited 2022 Aug 5];9:258. Available from: / pmc / articles / PMC2464610 / 12. Pineiro-Yanez E, Reboiro-Jato M, Gomez-Lopez G, Perales-Paton J, Troule K, Rodriguez JM, et al. PanDrugs: A novel method to prioritize anticancer drug treatments according to individual genomic data. Genome Med [Internet]. 2018 [cited 2020 Jul 7];10:41. Available from: https: / / genomemedicine.biomedcentral.com / articles / 10.1186 / s13073-018-0546-1 13. Huttlin EL, Bruckner RJ, Navarrete-Perea J, Cannon JR, Baltier K, Gebreab F, et al. Dual Proteome-scale Networks Reveal Cell-specific Remodeling of the Human Interactome. bioRxiv [Internet]. 2020 [cited 2020 Jul 7];2020.01.19.905109. Available from: https: / / www.biorxiv.org / content / biorxiv / early / 2020 / 01 / 19 / 2020.01.19.905109.full.pdf 14. Cheng F, Lu W, Liu C, Fang J, Hou Y, Handy DE, et al. A genome-wide positioning systems network algorithm for in silico drug repurposing. Nat Commun [Internet]. 2019 [cited 2020 Jul 19];10:1-14. Available from: https: / / doi.org / 10.1038 / s41467-019-10744-6 15. Luck K, Kim DK, Lambourne L, Spirohn K, Begg BE, Bian W, et al. A reference map of the human binary protein interactome. Nat 2020 5807803 [Internet]. 2020 [cited 2022 Aug 9];580:402-8. Available from: https: / / www.nature.com / articles / s41586-020-2188-x 16. Li T, Wernersson R, Hansen RB, Horn H, Mercer J, Slodkowicz G, et al. A scored human protein-protein interaction network to catalyze genomic interpretation. Nat Methods [Internet]. 2017 [cited 2018 Mar 11];14:61-4. Available from: http: / / www.nature.com / articles / nmeth.4083 17. Turner B, Razick S, Turinsky AL, Vlasblom J, Crowdy EK, Cho E, et al. iRefWeb: interactive analysis of consolidated protein interaction data and their supporting evidence. Database [Internet]. 2010 [cited 2022 Aug 9];2010. Available from: https: / / academic.oup.com / database / article / doi / 10.1093 / database / baq023 / 407721 18. Cerami EG, Gross BE, Demir E, Rodchenkov I, Babur O, Anwar N, et al. Pathway Commons, a web resource for biological pathway data. Nucleic Acids Res [Internet]. 2011 [cited 2019 Jan 28];39:D685-90. Available from: http: / / www.ncbi.nlm.nih.gov / pubmed / 21071392 19. Szklarczyk D, Gable AL, Lyon D, Junge A, Wyder S, Huerta-Cepas J, et al. STRING v11: protein-protein association networks with increased coverage, supporting functional discovery in genome-wide experimental datasets. Nucleic Acids Res [Internet]. 2019 [cited 2019 Jun 14];47:D607-13. Available from: https: / / academic.oup.com / nar / article / 47 / D1 / D607 / 5198476 20. Wainberg M, Kamber RA, Balsubramani A, Meyers RM, Sinnott-Armstrong N, Hornburg D, et al. A genome-wide atlas of co-essential modules assigns function to uncharacterized genes. Nat Genet 2021 535 [Internet]. 2021 [cited 2022 Jun 16];53:638-49. Available from: https: / / www.nature.com / articles / s41588-021-00840-z 21. Lee JS, Das A, Jerby-Arnon L, Arafeh R, Auslander N, Davidson M, et al. Harnessing synthetic lethality to predict the response to cancer treatment. Nat Commun [Internet]. 2018;9:1-12. Available from: http: / / dx.doi.org / 10.1038 / s41467-018-04647-1 22. Leiserson MDM, Vandin F, Wu HT, Dobson JR, Eldridge JV, Thomas JL, et al. Pan-cancer network analysis identifies combinations of rare somatic mutations across pathways and protein complexes. Nat Genet [Internet]. 2015 [cited 2018 Nov 27];47:106-14. Available from: http: / / www.nature.com / articles / ng.3168 23. Hristov BH, Chazelle B, Singh M. uKIN Combines New and Prior Information with Guided Network Propagation to Accurately Identify Disease Genes. Cell Syst [Internet]. 2020 [cited 2020 Aug 11];10:470-479.e3. Available from: https: / / doi.org / 10.1016 / j.cels.2020.05.008 [Explanation of symbols]

[0094] 27: New anticancer drug delivery device 2710: Data acquisition unit 2720: Network Production Department 2730: Module extraction unit 2740: Anticancer Drug Development Department

Claims

1. A method for deriving a new anticancer drug using a co-essentiality network by a computer device, (1) obtaining gene-genomic data, measuring the similarity between genes in the gene-genomic data, and creating a co-essentiality network; (2) extracting cancer-related driver modules from the co-essentiality network; (3) deriving the novel anticancer drug using the cancer-related driver module; The step (1) calculates a Pearson Correlation Coefficient (PCC) for each gene pair in the gene-genomic data; Applying a CLR (Context Likelihood Relatedness) algorithm to the absolute value of the PCC to measure the similarity; The method includes performing network propagation in which the step (2) uses a page-rank algorithm to prioritize genes in the co-essentiality network in order of their association with cancer-related driver genes.

2. The method of claim 1 , wherein the genetic and genomic data in step (1) is growth data of a cell line with loss of function of a gene.

3. 2. The method of claim 1, wherein step (2) comprises using the network propagation scores obtained through the network propagation to identify biological pathways associated with the cancer-associated driver genes.

4. In step (2), biological pathways that satisfy FDR (False Discovery Rate) < 0.01 and NES (Normalized Enrichment Score) > 0 are extracted from the identified biological pathways through GSEA (Gene Set Enrichment Analysis); The method of claim 3 , comprising selecting the biological pathway as the cancer-associated driver module.

5. The method of claim 4, wherein the cancer-related driver module is selected from the extracted biological pathways by a biological pathway having the lowest p-value through a log-rank test.

6. The method of claim 1 , wherein the cancer-associated driver module comprises the cancer-associated driver gene in the co-essentiality network.

7. The method of claim 1, wherein the novel anticancer drug comprises a repurposing of an existing drug.

8. A device for deriving a new anticancer drug using a co-essentiality network by a computer device, a data acquisition unit that acquires gene genome data; a network creation unit that creates a co-essentiality network by measuring the similarity between genes in the gene genome data; a module extraction unit that extracts a cancer-related driver module from the co-essentiality network; an anticancer drug derivation unit that derives the new anticancer drug using the cancer-related driver module; Including, The network creation unit calculates a Pearson Correlation Coefficient (PCC) for a gene pair in the gene genome data; Applying a CLR (Context Likelihood Relatedness) algorithm to the absolute value of the PCC to measure the similarity; The module extraction unit performs network propagation using a page-rank algorithm to prioritize genes in the co-essentiality network in order of their association with cancer-related driver genes.

9. The device according to claim 8 , wherein the genetic and genomic data of the data acquisition unit is growth data of a cell line due to loss of function of a gene.

10. The device of claim 8 , wherein the module extractor identifies a biological pathway associated with the cancer-related driver gene using a network propagation score obtained through the network propagation.

11. The module extraction unit extracts biological pathways that satisfy a False Discovery Rate (FDR)<0.01 and a Normalized Enrichment Score (NES)>0 from the identified biological pathways through Gene Set Enrichment Analysis (GSEA); The apparatus of claim 10 , wherein the biological pathway is selected as the cancer-associated driver module.

12. The device according to claim 11, wherein the cancer-related driver module is selected from the extracted biological pathways by a biological pathway having the lowest p-value through a log-rank test.

13. The apparatus of claim 8 , wherein the cancer-associated driver module includes the cancer-associated driver gene in the co-essentiality network.

14. The device according to claim 8 , wherein the novel anticancer drug comprises a repurposing of an existing drug.

Images