Table 3.

Overview of the different approaches to resolving problems in the analysis and treatment of ovarian cancer using modern technologies included in this paper. MS—mass spectrometry; miRNA—micro RNA, gDNA—genomic DNA, mRNA—messenger RNA, lncRNA—long non-coding RNA; LC-ESI-MS/MS—liquid chromatography-electrospray ionization/multi-stage mass spectrometry; LC-MS—liquid chromatography-mass spectrometry; RT-PCR—real-time PCR; DNMT—DNA methyltransferase; SNV—single nucleotide variation; CNV—copy number variation;

Problems	Approach	Method	Expected Application	Example Studies
Heterogeneity	Studying cell population patterns between ovarian cancer tumours of different grade, as well as between primary and metastatic tumours	Single-cell RNA sequencing	Understanding the leading cell population; may conclude in finding a specific target for diagnosis and precise treatment	[60]
	Proteomic profiling and statistical comparison between ovarian cancer cells and controls	Single-run MS	Potential biomarkers for diagnosis or outcome prediction	[80]
Late diagnosis	Training of machine to become a neural network with the lowest number of miRNAs needed for best diagnosis by correlation with clinical data	Machine learning algorithm based on miRNA expression data (microarrays, RNA sequencing)	Building of sensitive non-invasive diagnostic tools	[56]
	Using the physicochemical properties between alterations in genome methylation and gold surface	gDNA isolation and DNA–gold affinity	Development of easy, fast, and non-invasive diagnostic tools	[65]
Drug resistance	Building of endogenous RNA network	Support vector machine classifier (using data of mRNA, miRNA, and lncRNA vs. clinical data)	Development of a good model to predict disease reoccurrence in advance and to find potential biomarkers for the development of drug resistance	[52]
	Proteomic and metabolomics investigation and further statistical analysis to recognise differences between controls, platinum-resistant tumour, and platinum-sensitive tumour	2D-LC-ESI-MS/MS, LC-MS	Development of biomarkers for recognition of chemoresistant ovarian cancer	[81]
	Comparison of the primary sensitive and refractory resistant tumour	Whole-genome sequencing; transcriptome, methylation, and microRNA (miRNA) expression analyses	Designing of novel drugs for resensitisation or targeted therapy	[27]
Metastasis	Phylogenetic analyses identifying constituent clones and quantifying their relative abundances at multiple intraperitoneal sites	Whole-genome and single-nucleus sequencing	Understanding the process of metastasis migration and understanding the population spread, which could lead to better treatment management in the future	[46]
	Comparison of the mutation landscape, and copy number analysis between primary and metastatic sites	High-depth whole-exome sequencing	Understanding the ways of genomic evolution in transcoelomic metastasis	[45]
	Establishment, isolation, cloning, and propagation of the cellular content of ovarian multilayered spheroids (cancer stem cells) to study their clonogenic, tumourigenic, and invasive properties	In vitro and in vivo study, RT-PCR	Describing cellular mechanisms and the influence of cancer stem cells on the aggressiveness of ovarian cancer	[63]
Targeting	Treatment of heavily pretreated and chemoresistant patients with the addition of DNMT inhibitor	Clinical trial	Development of treatment which helps to restore the sensitivity to carboplatin (classic treatment)	[70,71]
	Finding SNV, CNV, alteration in mRNA expression, miRNA expression	Exome sequencing, RNA sequencing, integrated data analysis	Finding driver mutations and key disrupted pathways in pathogenesis for precision medicine	[26,49]
	Analysis of copy number signatures (including many copy number features)	Shallow whole-genome sequencing	Finding ways to predict overall survival and the probability of drug-resistance and relapse at the point of diagnosis	[28]
	10-mRNA-score model constructed so that it strongly correlates with the level of DNA mutations and predicts the genome instability	Construction of RNA network	Prediction model of poor outcome, which could identify important pathways for targeting disease	[55]