Multi-omics association study identifies new potential prostate cancer- causing gene

doi:10.21203/rs.3.rs-4663599/v1

Download PDF

Article

Multi-omics association study identifies new potential prostate cancer- causing gene

https://doi.org/10.21203/rs.3.rs-4663599/v1

This work is licensed under a CC BY 4.0 License

You are reading this latest preprint version

BACKGROUND

Prostate cancer is one of the most common malignant tumors among men worldwide, and surgery remains its mainstay of treatment. It is unclear how prostate cancer develops and what the most effective drug targets are for treating prostate cancer. Therefore, we sought to identify the genes responsible for the pathogenesis of prostate cancer.

METHODS

By integrating multidimensional and high-throughput data, proteome wide association studies (PWAS), transcriptome wide association studies (TWAS), single-cell sequencing, functional enrichment, Mendelian randomization (MR), and Bayesian co-localization analyses were used to screen for candidate genes that may contribute to prostate cancer and associate with clinical results of prostate cancer.

RESULTS

Our comprehensive analysis showed that protein abundance of eight genes was associated with prostate cancer, four of which were validated at the transcriptome level. These 8 candidate genes (MSMB, PLG, CHMP2B, ATF6B, EGF, TAPBP, GAS1 and MMP7) were validated. After combining single-cell sequencing, Mendelian randomization, and Bayesian co-localization analyses, we identified 1 gene (TAPBP) that is strongly associated with prostate cancer and explored the potential mechanisms of its pathogenicity and its clinical relevance.

Biological sciences/Cancer

Health sciences/Oncology

Health sciences/Urology

prostate cancer

Proteome‑wide association studies (PWAS)

Genome-wide association study (GWAS)

Transcriptome-wide association studies (TWAS)

Mendelian Randomization (MR) analysis

Prostate cancer (PCa) is one of the most common malignancies among men worldwide. According to Global Cancer Statistics 2020, PCa is the leading cause of cancer-related death in men [1]. Numerous PCa-associated risk loci have been screened using the genome-wide association study (GWAS) approach [2]. GWAS aims to tightly link genetic loci to diseases and other genetic traits [3, 4]. Over the past decade, the approach has involved many variant-phenotype associations [5] and has driven important scientific discoveries [6].

Despite the tremendous impact of GWAS, inherent difficulties continue to limit its success [7, 8]. An improvement of these methods would be a protein-centered approach that considers the impact of genetic variation on gene function. Proteins are the most effective biomarkers and therapeutic targets [9, 10] because they represent the major functional components of cellular and biological processes as well as the end products of gene expression [11]. The study of risk proteins in prostate cancer is crucial. Overall, studies have shown that alterations in proteomics are associated with cell cycle control, DNA repair, proteasome degradation and metabolic activity. In addition to differences in study methodology, possibly due to the heterogeneity of prostate cancer itself, the results are rather inconsistent [12].

Based on the available literature, we used proteome wide association studies (PWAS). PWAS is based on the premise that causal variation in coding regions affects phenotype by altering the biochemical function of a gene's protein product. To capture these effects, PWAS quantifies the extent to which proteins are impaired under individual genotypes.

Like other gene-based approaches, PWAS alleviates the burden of multiple test corrections. In addition, it provides specific functional explanations for the protein-coding genes it discovers. Therefore, we sought to identify new drug targets for the treatment of prostate cancer by combining high-throughput proteomics in prostate cancer with genetic data to determine the levels of genomic structure-associated proteins. To identify potential protein biomarkers, we took a six-step approach to systematically link protein biomarkers to prostate cancer. First, we used findings from the Atherosclerosis Risk in Communities (ARIC) dataset and prostate cancer GWAs to perform PWAS and its FDR-corrected analysis. Next, we analyzed the expression of PWAS-identified risk genes by the Prostate Cancer Single Cell Database and explored important genes driving PWAS signaling at the transcriptional level by utilizing gene expression data. Then we used independent Mendelian randomization (MR) analysis to validate PWAS significant genes and used COLOC to integrate the GWAs data with the prostate cancer database using Bayesian co-localization analysis to explore whether the two correlated signals were consistent with a common dependent variable(s). Fourth, we explored possible involvement in biological processes, cellular components, and molecular functions of TAPBP by enrichment analysis. Fifth, we explored the mutation and drug sensitivity of TAPBP. Finally, we explored the association of these potential prostate cancer-causing proteins with some prostate cancer clinical features.

Source of GWAS data

A large number of meta-analyses have been conducted on PCa GWAS, including the UK biobank (UKB) data are fully available at http://fastgwa.info/ukbimpbin and the GWAS Catalog. According to UKB and GWAS Catalog summary statistics, 7769 cases and 201039 controls were considered eligible. All participants for PCa included were of European descent.

Human blood proteomic and transcriptomic data

Serum proteomic data were collected Atherosclerosis Risk in Communities (ARIC) study; N ~ 9,000)[ 13] in a large population-based study. The ARIC study enrolled 15,792 participants from four American communities. The current study ultimately included 9,084 participants with plasma protein data. A total of 4,657 human serum proteins were considered eligible.

Local gene expression regulation is often shared across tissues [14, 15]and combining eQTL data across multiple tissues can improve the power of TWAS. Thus, Genotype-Tissue Expression (GTEx) version 8 data (n = 948) was used for the analysis of whole blood eQTLs [16]. A paired RNA-seq (Illumina TruSeq; Illumina Inc) was used to obtain gene expression data, and whole genome sequencing was used to obtain genotype data. The GTEx website provides information on donor registration, consent procedures, methods for obtaining biological samples, sample attachments, and procedures for histopathological examinations [17].

Single cell data

Single cell datasets of prostate cancer were downloaded from GEO with the access number of GSE141445. The pre-processing standards for Seurat object were as follow: number of RNA count > = 1000, number of features > = 200, percent.mt < = 20, percent.rb < = 20. Double cells were removed by R package DoubletFinder with default parameters.

Statistical analysis

Proteome‑wide association studies (PWAS)

The overall design of this study is shown in Fig. 1. The effect of SNPs on protein abundance was calculated for proteins with significant heritability (P < 0.01) using FUSION [18]. The common prediction models used in the analysis are top1, blup, lasso, enet and bslmmv [18]. the ARIC database was selected for this analysis, so only enet and top1 were selected for the analysis. [13]. To combine the genetic effects of PCa with protein weights for PWAS of PCa, a linear sum of Z-scores + weights of independent SNPs was calculated. To reduce the likelihood of false positives, a Bonferroni-corrected P-value threshold was employed. Benjamini-Hochberg (BH) was also used to impute the P value adjusted for false discovery rate (FDR).

Single cell analysis

R package harmony was applied to remove batch effect of samples from different patients. Cell type identification was determined by known marker genes and aid of R package SingleR. Enrichment score of GWAS related signatures for each cell were analyzed by function AddModuleScore from R package Seurat. More information about scRNA sequencing library preparation and sequence strategy can be found in work from study. In addition, more detailed parameters and pipelines for single cell analysis could refer to our previous works [19, 20, 21]

Transcriptome-wide association studies (TWAS)

The genetic influence of PCa was integrated with mRNA expression weights by using FUSION [18]. The basic procedure is as follows: Firstly, TWAS expression weights (i.e. SNP-gene correlations) are calculated from a reference expression panel [22]. To determine the out-of-sample R2 of each gene prediction model, FUSION performed five-fold cross-validation [18].

Mendelian Randomization (MR) analysis

The PWAS significant genes (from the FUSION method) were examined for cis-regulatory protein abundance using MR. The genome-wide significant (P < 5 × 10^− 6) SNPs were considered as inclusion and followed by linkage disequilibrium (LD) to screen for independent SNPs (R2 < 0.1). A harmonization of exposure (QTL) and outcome (PCa GWAS) data was then performed. Wald ratios can be used to estimate causality when only one independent QTL is available. In cases where multiple SNPs are available, the ratios of SNP exposures to SNP outcomes were estimated using the inverse variance weighting (IVW) method for random-effects meta-analysis. In addition, when the number of SNPs exceeded three, horizontal pleiotropy was tested using the MR-egger method. Bonferroni correction thresholds for the number of genes analyzed were set at P < 0.05/multiple comparisons [23, 24]. Mendelian randomization analysis was conducted using "TwoSampleMR" version 0.5.5 in R version 4.0.

Bayesian colocalization analysis

The Coloc Bayesian test was used to determine the probability that the PCa risk locus and the pQTL share the same causal signal [25, 26]. In prostate disease, p1 represents the probability that a specific variant correlates with a significant pQTL, p2 represents the likelihood that a particular variant will be associated with one in the disease. An individual variant's p12 is the probability of it being both an outcome and a pQTL in prostate disease. Five mutually exclusive hypotheses were tested: H0, no relation with either GWAS or pQTL; H1, relation with GWAS and no relation with pQTL; H2, relation with pQTL and no relation with GWAS; H3, relation with GWAS and pQTL, two independent SNPs; and H4, association with GWAS and pQTL, one common SNP. We focus on the last hypothesis, H4, and the posterior probability (PP) is employed to support H4 (denoted as PPH4). We define strong proof for co-localization when PPH4 ≥ 0.75.

Enrichment analysis based on discovery PWAS genes

Firstly, bulk sequencing datasets including ICGC-PRAD, TCGA-PRAD, GSE21034, GSE54460, GSE107299, GSE70768, GSE70769, GSE21034, DFZ2018 were downloaded from with the use of R packages TCGAbiolinks and GEOquary [27]. All transcriptome or microarray matrix were scaled and normalized before formal analysis. In addition, patients were excluded if missing clinical outcome information. Next, an enrichment score for each patient were calculated by R package GSVA with the function of ssGSEA [28]. Next, the biological annotation and hallmarks enrichment score correlated with score of discovery PWAS genes were analyzed by R packages clusterprofile and ggpubr, which was verified by algorithm of rank-based analysis (ORA).

Analysis of drug sensitivity analysis and mutation profile

Three drug sensitivity datasets, GDSC, CTRR and PRISM, were enrolled to detect the influence of GWAS score on therapy response or resistance by R package pRRophetic [29, 30].

All visualization were performed with the application of software including R and GraphPad. This study examined the clinical disparities of GSWA score using student T and Wilcoxon tests. The R package ggpubr was employed to assess the different GWAS score across different T, Stage, and Grade classifications through the Kruskal-Wallis test.

Discovery of PWAS in prostate cancer

The FUSION pipeline was used to integrate the PCa GWAS findings with human blood proteomes for the PWAS of PCa. Based on the 9084-master analyzed sample size derived from the ARIC database, the PWAS identified 16 genes (MSMB, PLG, CHMP2B, ATF6B, EGF, TAPBP, GAS1, MMP7, SERPINA3, AIF1, PRDX3, DKK3, PSAPL1, MINPP1, ANGPTL4, and CTSS), whose protein levels were associated with prostate cancer at P < 0.05 (Fig. 2). After Bonferroni correction threshold of P < 0.05/number, where Microseminoprotein-beta (MSMB), Plasminogen (Plg), CHMP2B, Activating transcription factor 6 (ATF6), Epidermal Growth Factor (EGF), Tapasin Binding Protein (TAPBP), Growth Arrest Specific Gene 1 (GAS1) and matrix metalloproteinase 7 (MMP7) were more significantly associated with prostate cancer.

Expression of these genes in different cell types of PCa

We investigated whether risk genes identified by PWAS are enriched in specific prostate cell types. Using human single-cell RNA-seq data from a cell-type database, we identified enriched causal genes expressed specifically in eight cell types (Fig. 3). TAPBP, ATF6B, CHMP2B, MSMB were expressed in a higher proportion of malignant cells, while TAPBP was expressed in a higher proportion and level in various immune cells.

TWAS identified four genes associated with PCa

We combined prostate cancer GWAS data with the human prostate transcriptome to perform transcriptome-wide association analysis (TWAS) of prostate cancer using FUSION. Expression of 4 out of 8 genes (CHMP2B, ATF6B, TAPBP, GAS1) in blood was associated with PCa (Table 1). This suggests that the combined evidence from PWAS and TWAS suggests a role in the pathogenesis of prostate cancer.

Table 1

TWAS identified four genes associated with Pca
gene	P value
CHMP2B	3.66E-04
ATF6B	1.68E-04
TAPBP	3.72E-03
GAS1	3.85E-01

8 genes associated with prostate cancer were validated by MR using prostate pQTL.

Most of the analyzed proteins could only be detected using a single SNP; therefore, MR estimation was mainly based on the Wald ratio method. We further identified the above eight proteins, and these biomarkers revealed significant evidence of association in prostate cancer GWAS (Table 2). We further analyzed their odds ratios and showed that among them, ATF6B, CHMP2B, GAS1, and MMP7 were the risk factors for the occurrence of PCa, while EGF, MSMB, PLG, TAPBP were protective factors (Fig. 4).

Table 2

Candidate genes identified by Mendelian randomization.
Gene	nsnps	IVW				MR Egger pval	Weighted median pval
Gene	nsnps	OR	ORLower	ORUpper	Pval	MR Egger pval	Weighted median pval
ATF6B	11	1.2966	1.0964	1.5334	0.002397	0.630797082	0.000124905
CHMP2B	21	1.1427	1.0800	1.2091	3.63E-06	0.160124428	2.18935E-05
EGF	12	0.8494	0.7963	0.9061	7.21E-07	0.020274016	2.29141E-05
GAS1	4	1.2076	1.0588	1.3773	0.004923	0.774952735	0.096094973
MMP7	20	1.0962	1.0434	1.1517	0.000265	0.49333611	0.010928317
MSMB	7	0.7767	0.6945	0.8688	9.74E-06	0.327374195	0.000347019
PLG	23	0.8492	0.8066	0.8940	4.61E-10	0.017827813	2.70502E-07
TAPBP	24	0.8942	0.8620	0.9276	2.35E-09	0.004858965	1.9132E-07

Co-localization between prostate cancer risk genes and pQTL in the prostate gland

Prostate cancer PWAS associations may arise from coincidental overlap between pQTL and loci that are in linkage disequilibrium with prostate cancer GWAS loci, or from the concurrent occurrence of a variant associated with protein expression, which is a protein quantitative trait locus (pQTL), and PCa. Statistical co-localization analyses for each gene report the probability of GWAS and pQTL sharing a causal variant, referred to as Hypothesis 4 (H4) and PP4/(PP3 + PP4) ≥ 0.75. Based on H4 ≥ 75% and PP4/(PP3 + PP4) ≥ 0.75, the analysis revealed three of the eight genes that provided evidence of genetic co-localization (MSMB, TAPBP, and EGF) (Table 3). This suggests that these three proteins play an important role in the pathophysiology of prostate cancer. TAPBP showed positive results in all of the above analyses, so its relationship with PCa needs to be further explored.

Table 3

Candidate genes identified by Bayesian colocalization.
Gene	nsnps	PP.H3.abf	PP.H4.abf
MSMB	596	0.000521504	0.99947791
TAPBP	5290	0.022416109	0.96049176
EGF	2841	0.079297104	0.88166609
CHMP2B	3177	0.99938787	0.00060374
GAS1	3039	0.063961688	0.46131011
MMP7	2986	0.39553352	0.02078762
PLG	811	0.994315549	0.00565747

Enrichment Analysis of the genes in PCa

In addition, we utilized the overrepresentation analysis (ORA) algorithm to decipher the role of TAPBP across different PCa cohorts. The GO results illustrated that TAPBP participated in cell adhesion, response to stimulus and nuclear division in BP; extracellular exosome, extracellular organelle and centromeric region in CC; and signaling receptor binding, protein − containing complex binding, and cell adhesion molecule binding in MF (Fig. 5A). KEGG analysis revealed that TAPBP might be involved in cell cycle, and p53 signaling pathway (Fig. 5B). GSEA and hallmark analysis also confirmed that TAPBP mainly participated in immune response, antigen processing and presentation, T cell receptor signaling pathway and Tnfa signaling via nfkb (Fig. 5C, D). Through the above finding, we found that TAPBP were involved in the immune response and cell cycle.

Mutation landscape between subtypes

The detailed genomic landscape difference between subgroups is depicted in Fig. 6A,

which indicates that the low TAPBP led to a high mutation frequency of speckle type BTB/POZ protein, SPOP, gain of 8q24.21 and 11q13.2 chromosome, and loss of 19q13.2 and 19q13.2 chromosome.

Drug sensitivity analysis

We harnessed three comprehensive drug sensitivity databases, GDSC, CTRP, and PRISM, to detect the relationship between TAPBP expression and therapy sensitivity or resistance. As Fig. 6B illustrated, high FDX1 expression displayed a consensus result of therapy resistance to Amuvatinib, Trichostatin, Pilaralisib, EHT − 1864, and Panobinostat in the GDSC database; MK − 2206, neuronal differentiation inducer III, lomeguatrib, BRD − A02303741:carboplatin (1:1 mol/mol), and necrostatin − 7 in the CTRP database; broxaterol, albuterol, GSK1904529A, pimobendan, thiamphenicol, eprinomectin, etc., in the PRISM database.

Potential prostate cancer-causing proteins with clinical associations

In furtherance of the understanding of the clinical association of TAPBP, a number of specific prostate cancer clinical indicators were used to correlate studies with it. As the most commonly used PCa screening indicator, PSA was first used for analysis. We found that the expression levels of TAPBP were not significantly associated with normal PSA (Fig. 7A). Biopsy is the gold standard for PCa diagnosis, and Gleason score is usually used to determine the benignity or malignancy of biopsied tissue. Our study found that as the Gleason score increased, the expression level of TAPBP decreased (Fig. 7B). Correspondingly TAPBP were more expressed overall in normal tissues than in tumor tissues (Fig. 7C). Further analysis of the relationship between PCa TNM staging and the expression of TAPBP showed a negative correlation between the relative expression of TAPBP and TNM staging (Fig. 7D, E, F), and showed lower expression in metastatic tumors (Fig. 7G). Tumor mutational burden (TMB), as an emerging biomarker, has received increasing attention for its role in predicting the efficacy of tumor immunotherapy. High GWAS scores are associated with low TMB (Fig. 7H).

Identifying therapeutic targets for disease is an important goal of human genetics research and is particularly important for PCA. In this study, we used a series of analytical techniques to investigate the functional associations between protein biomarkers in the prostate and prostate cancer. We initially identified eight potential risk genes for prostate cancer after PWAS analysis and FDR correction. Subsequently, we performed single-cell analysis of these genes and showed that they were differentially expressed in prostate. Further transcriptome association analysis revealed that four of the eight genes (CHMP2B, ATF6B, TAPBP, GAS1) were associated with PCa. Our results were replicated in MR validation analyses and verified whether the genes were risk factors or protective factors in the pathogenesis of PCa, providing a higher level of confidence. In addition, we identified MSMB, TAPBP and EGF by co-localization. Interestingly TAPBP was validated in all of the above analyses and further enrichment analyses showed that it is primarily involved in immune response and cell cycle regulation. This gene may serve as promising targets for further mechanistic, therapeutic studies, and clinical guidance.

TAPBP gene is close to the MHC and encodes a molecule that is a member of the IgSF. Its product, Tapasin, is required for the association of the MHC class I heterodimer with the Tap transporter protein in the endoplasmic reticulum (ER) [31]. It forms part of the peptide loading complex, which is essential for the stable assembly of class I molecules with peptides prior to transport to the cell surface.The TAPBP gene is located in the extended filamentous MHC region between the BING1 and RGL2 loci [32] .

Our study has several strengths. First, PWAS for prostate cancer was performed using the largest and most comprehensive pooled statistics from the human proteome and the most recent GWAS for prostate cancer.Second, we validated the risk proteins using independent MR validation analyses. Third, based on Bayesian co-localization used to estimate the probability of observing two correlated signals at specific loci with common causal variants, we confirmed the causative proteins of PCA (ICA1L, CAND2, and ALDH2). Finally, we performed single-cell analyses as well as enrichment analyses of the genes identified by PWAS to further validate their expression in tissues and the molecular functions that the gene products may exercise, the cellular environments in which they reside, and the biological processes in which they are involved.

There are also some limitations to this study. First, the data set used in this study failed to incorporate the gene KLK3 encoding PSA. Second, including methylation data in the analysis could provide a more comprehensive picture of disease progression. Last, most of our studies were conducted in Europeans, but we should be careful to generalize these findings to other races. The complexity of PCA biology and the molecular mechanisms behind it can only be understood through functional genomics approaches and biological experiments.

In conclusion, the comprehensive analysis provides compelling evidence supporting a causal association of TAPBP and tis transcription and translation products with PCa. These findings reveal the involvement of TAPBP-mediated cellular immunity, antigen recognition and presentation in the underlying mechanisms of PCa. In addition, this study also confirmed a number of proteins that have been clearly implicated in prostate carcinogenesis (MMP-7, EGFR, ATF6) as well as identifying a number of new or poorly-documented proteins that have been implicated in prostate carcinogenesis(GAS1, CHMP2B, PLG).

Competing interests

The authors declare that they have no competing interests.

Ethics approval and consent to participate

The ethics of this study observe the Helsinki Declaration. Ethics approval was obtained from the Ethics Committee of Shanghai Changhai Hospital (No. CHEC2021-086). Human Ethics and Consent to Participate declarations: not applicable

Finding

This work was supported by the 234 Panfeng Program at Changhai Hospital (No. 2020YXK030).

Author Contribution

XLW and AMJ: analyzing, writing-original draft, review, and editing. JMS: grammar correction and manuscript review. LW and ZYL: topic guidance and manuscript review.

Data Availability

All data generated or analyzed during this study are included in this published article and its supplementary information files. Consult the corresponding author (ZYL) for raw data.

Sung H, Ferlay J, Siegel RL, et al. Global Cancer Statistics 2020: GLOBOCAN Estimates of Incidence and Mortality Worldwide for 36 Cancers in 185 Countries. CA Cancer J Clin. 2021;71(3):209–249.
Schumacher FR, Al Olama AA, Berndt SI, et al. Association analyses of more than 140,000 men identify 63 new prostate cancer susceptibility loci. Nat Genet. 2018;50(7):928–936.
Loh PR, Tucker G, Bulik-Sullivan BK, et al. Efficient Bayesian mixed-model analysis increases association power in large cohorts. Nat Genet. 2015;47(3):284–90.
Chang CC, Chow CC, Tellier LC, et al. Second-generation PLINK: rising to the challenge of larger and richer datasets. Gigascience. 2015; 4: 7.
Buniello A, MacArthur JAL, Cerezo M, et al. The NHGRI-EBI GWAS Catalog of published genome-wide association studies, targeted arrays and summary statistics 2019. Nucleic Acids Res. 2019;47(D1): D1005-D1012.
Visscher PM, Wray NR, Zhang Q, et al. 10 Years of GWAS Discovery: Biology, Function, and Translation. Am J Hum Genet. 2017;101(1):5–22.
Korte A, Farlow A. The advantages and limitations of trait analysis with GWAS: a review. Plant Methods. 2013; 9: 29.
Manolio TA, Collins FS, Cox NJ, et al. Finding the missing heritability of complex diseases. Nature. 2009;461(7265):747–53.
Sharma K, Schmitt S, Bergner CG, T et al. Cell type- and brain region-resolved mouse brain proteome. Nat Neurosci. 2015;18(12):1819–31.
Vogel C, Marcotte EM. Insights into the regulation of protein abundance from proteomic and transcriptomic analyses. Nat Rev Genet. 2012;13(4):227–32.
Rolland DCM, Basrur V, Jeon YK, et al. Functional proteogenomics reveals biomarkers and therapeutic targets in lymphomas. Proc Natl Acad Sci U S A. 2017;114(25):6581–6586.
Intasqui P, Bertolla RP, Sadi MV. Prostate cancer proteomics: clinically useful protein biomarkers and future perspectives. Expert Rev Proteomics. 2018;15(1):65–79.
Zhang J, Dutta D, Köttgen A, et al. Plasma proteome analyses in individuals of European and African ancestry identify cis-pQTLs and models for proteome-wide association studies. Nat Genet. 2022;54(5):593–602.
GTEx Consortium. Human genomics. The Genotype-Tissue Expression (GTEx) pilot analysis: multitissue gene regulation in humans. Science. 2015;348(6235):648–60.
GTEx Consortium. Erratum: Genetic effects on gene expression across human tissues. Nature. 2018;553(7689):530.
Melé M, Ferreira PG, Reverter F, et al. The human transcriptome across tissues and individuals. Science. 2015;348(6235):660–5.
Nieuwenhuis TO, Rosenberg AZ, McCall MN, et al. Tissue, age, sex, and disease patterns of matrisome expression in GTEx transcriptome data. Sci Rep. 2021;11(1):21549.
Gusev A, Ko A, Shi H, et al. Integrative approaches for large-scale transcriptome-wide association studies. Nat Genet. 2016;48(3):245–52.
Jiang A, Zhou Y, Gong W, et al. CCNA2 as an Immunological Biomarker Encompassing Tumor Microenvironment and Therapeutic Response in Multiple Cancer Types. Oxid Med Cell Longev. 2022; 2022: 5910575.
Bao Y, Jiang A, Dong K, et al. DDX39 as a predictor of clinical prognosis and immune checkpoint therapy efficacy in patients with clear cell renal cell carcinoma. Int J Biol Sci. 2021;17(12):3158–3172.
Jiang A, Bao Y, Wang A, et al. Establishment of a Prognostic Prediction and Drug Selection Model for Patients with Clear Cell Renal Cell Carcinoma by Multiomics Data Analysis. Oxid Med Cell Longev. 2022; 2022: 3617775.
Fromer M, Roussos P, Sieberts SK, et al. Gene expression elucidates functional impact of polygenic risk for schizophrenia. Nat Neurosci. 2016;19(11):1442–1453.
Walker VM, Zheng J, Gaunt TR, Smith GD. Phenotypic Causal Inference Using Genome-Wide Association Study Data: Mendelian Randomization and Beyond. Annu Rev Biomed Data Sci. 2022; 5: 1–17.
Boehm FJ, Zhou X. Statistical methods for Mendelian randomization in genome-wide association studies: A review. Comput Struct Biotechnol J. 2022; 20: 2338–2351.
Giambartolomei C, Vukcevic D, Schadt EE, et al. Bayesian test for colocalisation between pairs of genetic association studies using summary statistics. PLoS Genet. 2014;10(5): e1004383.
Schmidt AW, Kühnapfel A, Kirsten H, et al. Colocalization analysis of pancreas eQTLs with risk loci from alcoholic and novel non-alcoholic chronic pancreatitis GWAS suggests potential disease causing mechanisms. Pancreatology. 2022;22(4):449–456.
Colaprico A, Silva TC, Olsen C, et al. TCGAbiolinks: an R/Bioconductor package for integrative analysis of TCGA data. Nucleic Acids Res. 2016;44(8): e71.
Hänzelmann S, Castelo R, Guinney J. GSVA: gene set variation analysis for microarray and RNA-seq data. BMC Bioinformatics. 2013; 14: 7.
Cokelaer T, Chen E, Iorio F, et al. GDSCTools for mining pharmacogenomic interactions in cancer. Bioinformatics. 2018;34(7):1226–1228.
Geeleher P, Cox N, Huang RS. pRRophetic: an R package for prediction of clinical chemotherapeutic response from tumor gene expression levels. PLoS One. 2014;9(9):e107468.
Herberg JA, Sgouros J, Jones T, et al. Genomic analysis of the Tapasin gene, located close to the TAP loci in the MHC. Eur J Immunol. 1998;28(2):459–67.
Teng MS, Stephens R, Du Pasquier L, et al. A human TAPBP (TAPASIN)-related gene, TAPBP-R. Eur J Immunol. 2002;32(4):1059–68.

No competing interests reported.

Download PDF

Reviews received at journal
09 Sep, 2024
Reviewers agreed at journal
13 Aug, 2024
Reviewers invited by journal
12 Aug, 2024
Editor assigned by journal
12 Aug, 2024
Editor invited by journal
05 Aug, 2024
Submission checks completed at journal
30 Jul, 2024
First submitted to journal
30 Jun, 2024

You are reading this latest preprint version

Multi-omics association study identifies new potential prostate cancer- causing gene

Status:

Version 1

Abstract

BACKGROUND

METHODS

RESULTS

Figures

Introduction

Method

Source of GWAS data

Human blood proteomic and transcriptomic data

Single cell data

Statistical analysis

Proteome‑wide association studies (PWAS)

Single cell analysis

Transcriptome-wide association studies (TWAS)

Mendelian Randomization (MR) analysis

Bayesian colocalization analysis

Enrichment analysis based on discovery PWAS genes

Analysis of drug sensitivity analysis and mutation profile

Results

Discovery of PWAS in prostate cancer

Expression of these genes in different cell types of PCa

TWAS identified four genes associated with PCa

Co-localization between prostate cancer risk genes and pQTL in the prostate gland

Enrichment Analysis of the genes in PCa

Mutation landscape between subtypes

Drug sensitivity analysis

Potential prostate cancer-causing proteins with clinical associations

Discussion

Conclusion

Declarations

Competing interests

Finding

Author Contribution

Data Availability

References

Additional Declarations

Supplementary Files

Status:

Version 1