Cancer Gene Regulatory Network Research Articles

Exploring gene regulatory interaction networks and predicting therapeutic molecules for hypopharyngeal cancer and EGFR-mutated lung adenocarcinoma.

Hypopharyngeal cancer gene regulatory networks and therapeutic molecules are a disease that is associated with EGFR-mutated lung adenocarcinoma. Here we utilized a bioinformatics approach to identify genetic commonalities between these two diseases. To this end, we examined microarray datasets from GEO (Gene Expression Omnibus) to identify differentially expressed genes, common genes, and hub genes between the selected two diseases. Our analyses identified potential therapeutic molecules for the selected diseases based on 10 hub genes with the highest interactions according to the degree topology method and the maximum clique centrality (MCC). These therapeutic molecules may have the potential for simultaneous treatment of these diseases.

Disruption of metazoan gene regulatory networks in cancer alters the balance of co-expression between genes of unicellular and multicellular origins

BackgroundMetazoans inherited genes from unicellular ancestors that perform essential biological processes such as cell division, metabolism, and protein translation. Multicellularity requires careful control and coordination of these unicellular genes to maintain tissue integrity and homeostasis. Gene regulatory networks (GRNs) that arose during metazoan evolution are frequently altered in cancer, resulting in over-expression of unicellular genes. We propose that an imbalance in co-expression of unicellular (UC) and multicellular (MC) genes is a driving force in cancer.ResultsWe combine gene co-expression analysis to infer changes to GRNs in cancer with protein sequence conservation data to distinguish genes with UC and MC origins. Co-expression networks created using RNA sequencing data from 31 tumor types and normal tissue samples are divided into modules enriched for UC genes, MC genes, or mixed UC-MC modules. The greatest differences between tumor and normal tissue co-expression networks occur within mixed UC-MC modules. MC and UC genes not commonly co-expressed in normal tissues form distinct co-expression modules seen only in tumors. The degree of rewiring of genes within mixed UC-MC modules increases with tumor grade and stage. Mixed UC-MC modules are enriched for somatic mutations in cancer genes, particularly amplifications, suggesting an important driver of the rewiring observed in tumors is copy number changes.ConclusionsOur study shows the greatest changes to gene co-expression patterns during tumor progression occur between genes of MC and UC origins, implicating the breakdown and rewiring of metazoan gene regulatory networks in cancer development and progression.

Advancing cancer driver gene detection via Schur complement graph augmentation and independent subspace feature extraction

Accurately identifying cancer driver genes (CDGs) is crucial for guiding cancer treatment and has recently received great attention from researchers. However, the high complexity and heterogeneity of cancer gene regulatory networks limit the precition accuracy of existing deep learning models. To address this, we introduce a model called SCIS-CDG that utilizes Schur complement graph augmentation and independent subspace feature extraction techniques to effectively predict potential CDGs. Firstly, a random Schur complement strategy is adopted to generate two augmented views of gene network within a graph contrastive learning framework. Rapid randomization of the random Schur complement strategy enhances the model's generalization and its ability to handle complex networks effectively. Upholding the Schur complement principle in expectations promotes the preservation of the original gene network's vital structure in the augmented views. Subsequently, we employ feature extraction technology using multiple independent subspaces, each trained with independent weights to reduce inter-subspace dependence and improve the model's expressiveness. Concurrently, we introduced a feature expansion component based on the structure of the gene network to address issues arising from the limited dimensionality of node features. Moreover, it can alleviate the challenges posed by the heterogeneity of cancer gene networks to some extent. Finally, we integrate a learnable attention weight mechanism into the graph neural network (GNN) encoder, utilizing feature expansion technology to optimize the significance of various feature levels in the prediction task. Following extensive experimental validation, the SCIS-CDG model has exhibited high efficiency in identifying known CDGs and uncovering potential unknown CDGs in external datasets. Particularly when compared to previous conventional GNN models, its performance has seen significant improved. The code and data are publicly available at: https://github.com/mxqmxqmxq/SCIS-CDG.

Lung Cancer Gene Regulatory Network of Transcription Factors Related to the Hallmarks of Cancer

The transcriptomic analysis of microarray and RNA-Seq datasets followed our own bioinformatic pipeline to identify a transcriptional regulatory network of lung cancer. Twenty-six transcription factors are dysregulated and co-expressed in most of the lung cancer and pulmonary arterial hypertension datasets, which makes them the most frequently dysregulated transcription factors. Co-expression, gene regulatory, coregulatory, and transcriptional regulatory networks, along with fibration symmetries, were constructed to identify common connection patterns, alignments, main regulators, and target genes in order to analyze transcription factor complex formation, as well as its synchronized co-expression patterns in every type of lung cancer. The regulatory function of the most frequently dysregulated transcription factors over lung cancer deregulated genes was validated with ChEA3 enrichment analysis. A Kaplan-Meier plotter analysis linked the dysregulation of the top transcription factors with lung cancer patients' survival. Our results indicate that lung cancer has unique and common deregulated genes and transcription factors with pulmonary arterial hypertension, co-expressed and regulated in a coordinated and cooperative manner by the transcriptional regulatory network that might be associated with critical biological processes and signaling pathways related to the acquisition of the hallmarks of cancer, making them potentially relevant tumor biomarkers for lung cancer early diagnosis and targets for the development of personalized therapies against lung cancer.

Identification of potential microRNAs regulating metabolic plasticity and cellular phenotypes in glioblastoma.

MicroRNAs (miRNAs) play important role in regulating cellular metabolism, and are currently being explored in cancer. As metabolic reprogramming in cancer is a major mediator of phenotypic plasticity, understanding miRNA-regulated metabolism will provide opportunities to identify miRNA targets that can regulate oncogenic phenotypes by taking control of cellular metabolism. In the present work, we studied the effect of differentially expressed miRNAs on metabolism, and associated oncogenic phenotypes in glioblastoma (GBM) using patient-derived data. Networks of differentially expressed miRNAs and metabolic genes were created and analyzed to identify important miRNAs that regulate major metabolism in GBM. Graph network-based approaches like network diffusion, backbone extraction, and different centrality measures were used to analyze these networks for identification of potential miRNA targets. Important metabolic processes and cellular phenotypes were annotated to trace the functional responses associated with these miRNA-regulated metabolic genes and associated phenotype networks. miRNA-regulated metabolic gene subnetworks of cellular phenotypes were extracted, and important miRNAs regulating these phenotypes were identified. The most important outcome of the study is the target miRNA combinations predicted for five different oncogenic phenotypes that can be tested experimentally for miRNA-based therapeutic design in GBM. Strategies implemented in the study can be used to generate testable hypotheses in other cancer types as well, and design context-specific miRNA-based therapy for individual patient. Their usability can be further extended to other gene regulatory networks in cancer and other genetic diseases.

GENAVOS: A New Tool for Modelling and Analyzing Cancer Gene Regulatory Networks Using Delayed Nonlinear Variable Order Fractional System

Gene regulatory networks (GRN) are one of the etiologies associated with cancer. Their dysregulation can be associated with cancer formation and asymmetric cellular functions in cancer stem cells, leading to disease persistence and resistance to treatment. Systems that model the complex dynamics of these networks along with adapting to partially known real omics data are closer to reality and may be useful to understand the mechanisms underlying neoplastic phenomena. In this paper, for the first time, modelling of GRNs is performed using delayed nonlinear variable order fractional (VOF) systems in the state space by a new tool called GENAVOS. Although the tool uses gene expression time series data to identify and optimize system parameters, it also models possible epigenetic signals, and the results show that the nonlinear VOF systems have very good flexibility in adapting to real data. We found that GRNs in cancer cells actually have a larger delay parameter than in normal cells. It is also possible to create weak chaotic, periodic, and quasi-periodic oscillations by changing the parameters. Chaos can be associated with the onset of cancer. Our findings indicate a profound effect of time-varying orders on these networks, which may be related to a type of cellular epigenetic memory. By changing the delay parameter and the variable order functions (possible epigenetics signals) for a normal cell system, its behaviour becomes quite similar to the behaviour of a cancer cell. This work confirms the effective role of the miR-17-92 cluster as an epigenetic factor in the cancer cell cycle.

Feature selection for data integration with mixed multiview data

Data integration methods that analyze multiple sources of data simultaneously can often provide more holistic insights than can separate inquiries of each data source. Motivated by the advantages of data integration in the era of “big data,” we investigate feature selection for high-dimensional multiview data with mixed data types (e.g., continuous, binary, count-valued). This heterogeneity of multiview data poses numerous challenges for existing feature selection methods. However, after critically examining these issues through empirical and theoretically-guided lenses, we develop a practical solution, the Block Randomized Adaptive Iterative Lasso (B-RAIL) which combines the strengths of the randomized Lasso, adaptive weighting schemes and stability selection. B-RAIL serves as a versatile data integration method for sparse regression and graph selection, and we demonstrate the effectiveness of B-RAIL through extensive simulations and a case study to infer the ovarian cancer gene regulatory network. In this case study, B-RAIL successfully identifies well-known biomarkers associated with ovarian cancer and hints at novel candidates for future ovarian cancer research.

NIMBus: a negative binomial regression based Integrative Method for mutation Burden Analysis

BackgroundIdentifying frequently mutated regions is a key approach to discover DNA elements influencing cancer progression.However, it is challenging to identify these burdened regions due to mutation rate heterogeneity across the genome and across different individuals. Moreover, it is known that this heterogeneity partially stems from genomic confounding factors, such as replication timing and chromatin organization. The increasing availability of cancer whole genome sequences and functional genomics data from the Encyclopedia of DNA Elements (ENCODE) may help address these issues.ResultsWe developed a negative binomial regression-based Integrative Method for mutation Burden analysiS (NIMBus). Our approach addresses the over-dispersion of mutation count statistics by (1) using a Gamma–Poisson mixture model to capture the mutation-rate heterogeneity across different individuals and (2) estimating regional background mutation rates by regressing the varying local mutation counts against genomic features extracted from ENCODE. We applied NIMBus to whole-genome cancer sequences from the PanCancer Analysis of Whole Genomes project (PCAWG) and other cohorts. It successfully identified well-known coding and noncoding drivers, such as TP53 and the TERT promoter. To further characterize the burdening of non-coding regions, we used NIMBus to screen transcription factor binding sites in promoter regions that intersect DNase I hypersensitive sites (DHSs). This analysis identified mutational hotspots that potentially disrupt gene regulatory networks in cancer. We also compare this method to other mutation burden analysis methods.ConclusionNIMBus is a powerful tool to identify mutational hotspots. The NIMBus software and results are available as an online resource at github.gersteinlab.org/nimbus.

Variant analysis of prostate cancer in Japanese patients and a new attempt to predict related biological pathways.

There are regional and/or ethnic differences in tumorigenic pathways among several types of cancer, including prostate cancer (PCa). However, information on genome‑wide gene alterations and the transcriptome is currently only available for PCa patients from Western countries. In order to profile the genetic alterations in Japanese patients with PCa, new panels were created to examine nucleotide sequence variations in 71 selected PCa‑related genes (KCC71) and to detect all fusion RNA transcripts known in PCa (PCaFusion). An analysis of 21 Japanese PCa cases identified 33 different somatic variants in 24genes in the KCC71 panel, including 2in SPOP (F102V and F133L), 2in BRCA2 (I1859fs and R2318ter, resulting in premature termination of the polypeptide), and 1each in BRAF (K601E), CDH1 (E880K) and RB1 (R621S), as pathogenic alterations. Unexpectedly, the TMPRSS2‑ERG fusion transcript was detected in only 1case, although the SLC45A3‑ELK4 and USP9Y‑TTTY15 fusion transcripts, known as transcription‑mediated chimeric RNAs, were detected in all examined cases. A new pathway analysis with The Cancer Network Galaxy (TCNG), a cancer gene regulatory network database, was also applied in an attempt to predict molecular pathways implicated in PCa in the Japanese population. Based on the 24genes having somatic variants identified by the panel analysis as initial seed genes, a putative core network was finally established, including 5 identified genes, namely TNK2, SOX9, CDH1, FOXA1 and TP53, with high commonality from TCNG datasets. These genes are expected to be involved in tumor development, as revealed by the results of an enrichment analysis with Gene Ontology terms. This analysis must be further extended to include more cases in order to verify this method and also to elucidate the characteristics of PCa in Japanese patients.

Reconciling periodic rhythms of large-scale biological networks by optimal control.

Periodic rhythms are ubiquitous phenomena that illuminate the underlying mechanism of cyclic activities in biological systems, which can be represented by cyclic attractors of the related biological network. Disorders of periodic rhythms are detrimental to the natural behaviours of living organisms. Previous studies have shown that the state transition from one to another attractor can be accomplished by regulating external signals. However, most of these studies until now have mainly focused on point attractors while ignoring cyclic ones. The aim of this study is to investigate an approach for reconciling abnormal periodic rhythms, such as diminished circadian amplitude and phase delay, to the regular rhythms of complex biological networks. For this purpose, we formulate and solve a mixed-integer nonlinear dynamic optimization problem simultaneously to identify regulation variables and to determine optimal control strategies for state transition and adjustment of periodic rhythms. Numerical experiments are implemented in three examples including a chaotic system, a mammalian circadian rhythm system and a gastric cancer gene regulatory network. The results show that regulating a small number of biochemical molecules in the network is sufficient to successfully drive the system to the target cyclic attractor by implementing an optimal control strategy.

Computational Inference of Gene Co-Expression Networks for the identification of Lung Carcinoma Biomarkers: An Ensemble Approach.

Gene Networks (GN), have emerged as an useful tool in recent years for the analysis of different diseases in the field of biomedicine. In particular, GNs have been widely applied for the study and analysis of different types of cancer. In this context, Lung carcinoma is among the most common cancer types and its short life expectancy is partly due to late diagnosis. For this reason, lung cancer biomarkers that can be easily measured are highly demanded in biomedical research. In this work, we present an application of gene co-expression networks in the modelling of lung cancer gene regulatory networks, which ultimately served to the discovery of new biomarkers. For this, a robust GN inference was performed from microarray data concomitantly using three different co-expression measures. Results identified a major cluster of genes involved in SRP-dependent co-translational protein target to membrane, as well as a set of 28 genes that were exclusively found in networks generated from cancer samples. Amongst potential biomarkers, genes and are highlighted due to their implications in a considerable portion of lung and bronchus primary carcinomas. These findings demonstrate the potential of GN reconstruction in the rational prediction of biomarkers.

Abstract SY21-01: Analyzing tumor metabolism in patients

Abstract Metabolism is dynamic and responds to a wide range of cell-intrinsic and cell-extrinsic factors. Many human diseases involve perturbations of metabolism at the cellular level, and in many cases normalizing the metabolic state has proven to be therapeutically valuable. Cancer is one such disease, in which factors both intrinsic and extrinsic to the malignant cell impact tumor biology and disease progression. In cancer, cell-intrinsic influences on metabolism include somatically-acquired mutations in oncogenes and tumor suppressor genes, many of which regulate metabolic activity. Cell-extrinsic factors include nutrient access, which may become limiting due to inadequate vasculature and intense fuel utilization, and metabolic interactions with stromal and immune cells. A major challenge in cancer metabolism research is to understand how these various factors culminate in the metabolic phenotype of an intact tumor, and ultimately to identify which altered pathways represent potential therapeutic targets. Among the many metabolic pathways reprogrammed in cancer, nucleotide metabolism and responses to DNA damage are highly integrated with aberrant signaling and gene regulatory networks in cancer. We have taken two parallel approaches to understand metabolic complexity and identify potential therapeutic liabilities in human cancer. The first uses a combination of multi-parametric imaging and intraoperative stable isotope infusions to assess metabolic fluxes in patients with solid tumors, and to compare fluxes between tumors and adjacent tissue. Genomic, histological and metabolic analysis of tumor samples allows us to correlate various intrinsic and extrinsic factors to specific aspects of the metabolic phenotype. The second approach uses standardized culture conditions to assess cell-intrinsic heterogeneity of metabolic preferences and dependencies in large panels of human cancer cell lines. This approach has uncovered liabilities associated with specific molecular subtypes of non-small cell and small cell lung cancerm ncluding several related to nucleotide and nucleic acid metabolism. I will discuss the application of these approaches to metabolic heterogeneity in human cancer, with an emphasis on features relevant to intact tumors and therapeutic liabilities and the role of metabolic imaging in studying cancer metabolism. Citation Format: Ralph J. Deberardinis. Analyzing tumor metabolism in patients [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2019; 2019 Mar 29-Apr 3; Atlanta, GA. Philadelphia (PA): AACR; Cancer Res 2019;79(13 Suppl):Abstract nr SY21-01.

Identification of regulatory variables for state transition of biological networks

Attractors represent steady states of biological networks. Recent studies have shown that regulatory variables can be used to steer a network state transition from an undesired attractor, such as a cancerous state, to a desired healthy one. Therefore, it is important to identify the regulatory variables and determine their time-dependent profile for state transition of a given network. However, this is a challenging task since regulatory variables have to be identified among numerous candidates in a large-scale biological network. In this study, we developed a new method for identifying regulatory variables in large-scale biological networks for the purpose of state transition. As a result, a set of optimal regulatory variables can be determined based on formulating and solving a mixed-integer nonlinear dynamic optimization problem. A relaxation scheme is used to overcome the difficulties in solving this complex problem containing a large number of binary variables. The solution to this problem simultaneously identifies the optimal regulatory variables, provides strength of regulatory interactions, and obtains the minimal control time to realize the required state transition. In addition, by adjusting the objective function, various combinations of the strength of regulatory interactions and the transition time can be achieved according to the requirement for disease therapy. Results of three case studies (a myeloid differentiation regulatory network, a cancer gene regulatory network, and a T-LGL signaling network) demonstrate the efficacy of the proposed approach. Therefore, this study establishes an appropriate framework for identifying the regulatory variables for state transition of complex biological networks.

Prostate Cancer Gene Regulatory Network Inferred from RNA-Seq Data.

Background: Cancer is a complex disease with a lucid etiology and in understanding the causation, we need to appreciate this complexity.Objective: Here we are aiming to gain insights into the genetic associations of prostate cancer through a network-based systems approach using the BC3Net algorithm.Methods: Specifically, we infer a prostate cancer Gene Regulatory Network (GRN) from a large-scale gene expression data set of 333 patient RNA-seq profiles obtained from The Cancer Genome Atlas (TCGA) database.Results: We analyze the functional components of the inferred network by extracting subnetworks based on biological process information and interpret the role of known cancer genes within each process. Fur-thermore, we investigate the local landscape of prostate cancer genes and discuss pathological associa-tions that may be relevant in the development of new targeted cancer therapies.Conclusion: Our network-based analysis provides a practical systems biology approach to reveal the collective gene-interactions of prostate cancer. This allows a close interpretation of biological activity in terms of the hallmarks of cancer.

Identification of cis-regulatory mutations generating de novo edges in personalized cancer gene regulatory networks

The identification of functional non-coding mutations is a key challenge in the field of genomics. Here we introduce μ-cisTarget to filter, annotate and prioritize cis-regulatory mutations based on their putative effect on the underlying “personal” gene regulatory network. We validated μ-cisTarget by re-analyzing the TAL1 and LMO1 enhancer mutations in T-ALL, and the TERT promoter mutation in melanoma. Next, we re-sequenced the full genomes of ten cancer cell lines and used matched transcriptome data and motif discovery to identify master regulators with de novo binding sites that result in the up-regulation of nearby oncogenic drivers. μ-cisTarget is available from http://mucistarget.aertslab.org.

Bioinformatic exploration of MTA1-regulated gene networks in colon cancer.

Metastasis-associated gene 1 (MTA1) controls a series of biological processes in tumor progression. Tumor progression is a complex process regulated by a gene network. The global cancer gene regulatory network must be analyzed to determine the position of MTA1 in the molecular network and its cooperative genes by further exploring the biological functions of this gene. We used TCGA data sets and GeneCards database to screen MTA1-related genes. GO and KEGG pathway analyses were conducted with DAVID and gene network analysis via STRING and Cytoscape. Results showed that in the development of colon cancer, MTA1 is linked to certain signal pathways, such as Wnt/Notch/nucleotide excision repair pathways. The findings also suggested that MTA1 demonstrates the closest relationship in a coregulation process with the key molecules AKT1, EP300, CREBBP, SMARCA4, RHOA, and CAD. These results lead MTA1 exploration to an in-depth investigation in different directions, such as Wnt, Notch, and DNA repair.

A physical mechanism of cancer heterogeneity.

We studied a core cancer gene regulatory network motif to uncover possible source of cancer heterogeneity from epigenetic sources. When the time scale of the protein regulation to the gene is faster compared to the protein synthesis and degradation (adiabatic regime), normal state, cancer state and an intermediate premalignant state emerge. Due to the epigenetics such as DNA methylation and histone remodification, the time scale of the protein regulation to the gene can be slower or comparable to the protein synthesis and degradation (non-adiabatic regime). In this case, many more states emerge as possible phenotype alternations. This gives the origin of the heterogeneity. The cancer heterogeneity is reflected from the emergence of more phenotypic states, larger protein concentration fluctuations, wider kinetic distributions and multiplicity of kinetic paths from normal to cancer state, higher energy cost per gene switching, and weaker stability.

Regulators of genetic risk of breast cancer identified by integrative network analysis

Genetic risk for breast cancer is conferred by a combination of multiple variants of small effect. To better understand how risk loci might combine, we examined whether risk-associated genes share regulatory mechanisms. We created a breast cancer gene regulatory network between transcription factors (TFs) and putative target genes (regulons) and asked whether specific regulons are enriched for genes associated with risk loci via eQTLs. We identified 36 overlapping regulons that were enriched and formed a distinct cluster within the network, suggesting shared biology. The risk-TFs driving these regulons are frequently mutated in cancer and lie in two opposing subgroups, which relate to ER+ luminal A/B and to ER− basal-like cancers and to different, luminal epithelial cell populations in the adult mammary gland. Our network approach provides a foundation to reveal the regulatory circuits governing breast cancer, to identify targets for intervention, and is transferable to other disease settings.

Urothelial cancer gene regulatory networks inferred from large-scale RNAseq, Bead and Oligo gene expression data.

BackgroundUrothelial pathogenesis is a complex process driven by an underlying network of interconnected genes. The identification of novel genomic target regions and gene targets that drive urothelial carcinogenesis is crucial in order to improve our current limited understanding of urothelial cancer (UC) on the molecular level. The inference of genome-wide gene regulatory networks (GRN) from large-scale gene expression data provides a promising approach for a detailed investigation of the underlying network structure associated to urothelial carcinogenesis.MethodsIn our study we inferred and compared three GRNs by the application of the BC3Net inference algorithm to large-scale transitional cell carcinoma gene expression data sets from Illumina RNAseq (179 samples), Illumina Bead arrays (165 samples) and Affymetrix Oligo microarrays (188 samples). We investigated the structural and functional properties of GRNs for the identification of molecular targets associated to urothelial cancer.ResultsWe found that the urothelial cancer (UC) GRNs show a significant enrichment of subnetworks that are associated with known cancer hallmarks including cell cycle, immune response, signaling, differentiation and translation. Interestingly, the most prominent subnetworks of co-located genes were found on chromosome regions 5q31.3 (RNAseq), 8q24.3 (Oligo) and 1q23.3 (Bead), which all represent known genomic regions frequently deregulated or aberated in urothelial cancer and other cancer types. Furthermore, the identified hub genes of the individual GRNs, e.g., HID1/DMC1 (tumor development), RNF17/TDRD4 (cancer antigen) and CYP4A11 (angiogenesis/ metastasis) are known cancer associated markers. The GRNs were highly dataset specific on the interaction level between individual genes, but showed large similarities on the biological function level represented by subnetworks. Remarkably, the RNAseq UC GRN showed twice the proportion of significant functional subnetworks. Based on our analysis of inferential and experimental networks the Bead UC GRN showed the lowest performance compared to the RNAseq and Oligo UC GRNs.ConclusionTo our knowledge, this is the first study investigating genome-scale UC GRNs. RNAseq based gene expression data is the data platform of choice for a GRN inference. Our study offers new avenues for the identification of novel putative diagnostic targets for subsequent studies in bladder tumors.Electronic supplementary materialThe online version of this article (doi:10.1186/s12918-015-0165-z) contains supplementary material, which is available to authorized users.

Quantifying the underlying landscape and paths of cancer.

Cancer is a disease regulated by the underlying gene networks. The emergence of normal and cancer states as well as the transformation between them can be thought of as a result of the gene network interactions and associated changes. We developed a global potential landscape and path framework to quantify cancer and associated processes. We constructed a cancer gene regulatory network based on the experimental evidences and uncovered the underlying landscape. The resulting tristable landscape characterizes important biological states: normal, cancer and apoptosis. The landscape topography in terms of barrier heights between stable state attractors quantifies the global stability of the cancer network system. We propose two mechanisms of cancerization: one is by the changes of landscape topography through the changes in regulation strengths of the gene networks. The other is by the fluctuations that help the system to go over the critical barrier at fixed landscape topography. The kinetic paths from least action principle quantify the transition processes among normal state, cancer state and apoptosis state. The kinetic rates provide the quantification of transition speeds among normal, cancer and apoptosis attractors. By the global sensitivity analysis of the gene network parameters on the landscape topography, we uncovered some key gene regulations determining the transitions between cancer and normal states. This can be used to guide the design of new anti-cancer tactics, through cocktail strategy of targeting multiple key regulation links simultaneously, for preventing cancer occurrence or transforming the early cancer state back to normal state.