Dataset Information

Discovering the molecular differences between right- and left-sided colon cancer using machine learning methods.

ABSTRACT: BACKGROUND:In recent years, the differences between left-sided colon cancer (LCC) and right-sided colon cancer (RCC) have received increasing attention due to the clinicopathological variation between them. However, some of these differences have remained unclear and conflicting results have been reported. METHODS:From The Cancer Genome Atlas (TCGA), we obtained RNA sequencing data and gene mutation data on 323 and 283 colon cancer patients, respectively. Differential analysis was firstly done on gene expression data and mutation data between LCC and RCC, separately. Machine learning (ML) methods were then used to select key genes or mutations as features to construct models to classify LCC and RCC patients. Finally, we conducted correlation analysis to identify the correlations between differentially expressed genes (DEGs) and mutations using logistic regression (LR) models. RESULTS:We found distinct gene mutation and expression patterns between LCC and RCC patients and further selected the 30 most important mutations and 17 most important gene expression features using ML methods. The classification models created using these features classified LCC and RCC patients with high accuracy (areas under the curve (AUC) of 0.8 and 0.96 for mutation and gene expression data, respectively). The expression of PRAC1 and BRAF V600E mutation (rs113488022) were the most important feature for each model. Correlations of mutations and gene expression data were also identified using LR models. Among them, rs113488022 was found to have significance relevance to the expression of four genes, and thus should be focused on in further study. CONCLUSIONS:On the basis of ML methods, we found some key molecular differences between LCC and RCC, which could differentiate these two groups of patients with high accuracy. These differences might be key factors behind the variation in clinical features between LCC and RCC and thus help to improve treatment, such as determining the appropriate therapy for patients.

SUBMITTER: Jiang Y

PROVIDER: S-EPMC7574488 | biostudies-literature | 2020 Oct

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Discovering the molecular differences between right- and left-sided colon cancer using machine learning methods.

Jiang Yimei Y Yan Xiaowei X Liu Kun K Shi Yiqing Y Wang Changgang C Hu Jiele J Li You Y Wu Qinghua Q Xiang Ming M Zhao Ren R

BMC cancer 20201019 1

<h4>Background</h4>In recent years, the differences between left-sided colon cancer (LCC) and right-sided colon cancer (RCC) have received increasing attention due to the clinicopathological variation between them. However, some of these differences have remained unclear and conflicting results have been reported.<h4>Methods</h4>From The Cancer Genome Atlas (TCGA), we obtained RNA sequencing data and gene mutation data on 323 and 283 colon cancer patients, respectively. Differential analysis was ...[more]

PMID: 33076847

Similar Datasets

Project description:The colon is derived from the embryological midgut and hindgut separately, with the right colon and left colon having different features with regards to both anatomical and physiological characteristics. Cancers located in the right and left colon are referred to as right colon cancer (RCC) and left colon cancer (LCC), respectively, based on their apparent anatomical positions. Increasing evidence supports the notion that not only are there differences in treatment strategies when dealing with RCC and LCC, but molecular features also vary between them, not to mention the distinguishing clinical manifestations. Disease-free survival after radical surgery of both RCC and LCC are similar. In the treatment of RCC, the benefit gained from adjuvant FOLFIRI chemotherapy is superior, or at least similar, to LCC, but inferior to LCC if FOLFOX regimen is applied. On the other hand, metastatic LCC exhibits longer survival than that of RCC in a palliative chemotherapy setting. For KRAS wild-type cancers, LCC benefits more from cetuximab treatment than RCC. Moreover, advanced LCC shows a higher sensitivity to bevacizumab treatment in comparison with advanced RCC. Significant varieties exist at the molecular level between RCC and LCC, which may serve as the cause of all apparent differences. With respect to carcinogenesis mechanisms, RCC is associated with known gene types, such as MMR, KRAS, BRAF, and miRNA-31, while LCC is associated with CIN, p53, NRAS, miRNA-146a, miRNA-147b, and miRNA-1288. Regarding protein expression, RCC is related to GNAS, NQO1, telomerase activity, P-PDH, and annexin A10, while LCC is related to Topo I, TS, and EGFR. In addition, separated pathways dominate progression to relapse in RCC and LCC. Therefore, RCC and LCC should be regarded as two heterogeneous entities, with this heterogeneity being used to stratify patients in order for them to have the optimal, current, and novel therapeutic strategies in clinical practice. Additional research is needed to uncover further differences between RCC and LCC.

Project description:The difference between left- and right-sided colon cancer has become the focus of global attention, and researchers have found differences in the morbidity, molecular biological characteristics, and response to targeted drug therapy between left- and right-sided colon cancer. Therefore, the identification of more effective predictive indicators is critical for providing guidance to future clinical work. We collected samples from different colon sites and regions and analyzed the identities and distributions of differentially expressed species in the microbiota in the left and right sides of the colon to better explore the pathogenesis of colon cancer and provided a basis for individualized drug therapy. We collected samples from different regions in the body of 40 patients with colon cancer, including stool and tissues. The Subjects were classified into four groups, and this classification was mainly based on the colon cancer distribution. The microbiota composition of the left-sided and right-sided colon samples was assessed by specifically amplifying the V3-V4 region of the 16S rDNA gene from DNA extracts from the samples. These amplicons were examined by Illumina HiSeq 2500 sequencing. The microbial taxa in the left-sided colon samples are more abundant than those in the right-sided colon samples. The flora in the left-sided colon samples, such as Clostridium perfringens and Fusobacterium nucleatum, might be associated with VEGF expression and are more likely to promote colon cancer. The microbiota distribution in the right-sided colon samples is less invasive and harmful and particularly rich in Bifidobacterium dentium. In addition, Streptococcus, which is the target of EGFR, was found to be expressed in both the left- and right-sided colon samples but was found at a higher level in the left-sided colon samples. Additionally, the differential pathways involved in the left-sided colon samples mainly mediate DNA damage, methylation, and histone modifications, whereas those in the right-sided colon samples are dominated by DNA synthesis. The comparison of only the geographical differences revealed a significant difference in the distribution of the microbial population. The adherent microbiota composition and structural changes between the left- and right-sided colon samples might contribute to the development of colon cancer, lead to different morbidities, and further affect the prognosis of patients and their sensitivity to targeted drugs. Therefore, the identification of the differential flora in the colon could be used as an indicator for predicting the occurrence and development of colon cancer, which is also beneficial for future individualized drug therapy.

Project description:BackgroundGiven the differences in embryonic origin, vascular and nervous supplies, microbiotic burden, and main physiological functions of left and right colons, tumor location is increasingly suggested to dictate tumor behavior affecting pathology, progression and prognosis. Right-sided colon cancers arise in the cecum, ascending colon, hepatic flexure and/or transverse colon, while left-sided colon cancers arise in the splenic flexure, descending, and/or sigmoid colon. In contrast to prior reports, we attempt to delineate programs of tumorigenesis independently for each side.MethodsFour hundred and eleven samples were extracted from The Cancer Genome Atlas-COAD cohort, based on a conservative sample inclusion criterion. Each side was independently analyzed with respect to their respective normal tissue, at the level of transcription, post-transcription, miRNA control and methylation in both a stage specific and stage-agnostic manner.ResultsOur results indicate a suppression of enzymes involved in various stages of carcinogen breakdown including CYP2C8, CYP4F12, GSTA1, and UGT1A within right colon tumors. This implies its reduced capacity to detoxify carcinogens, contributing to a genotoxic tumor environment, and subsequently a more aggressive phenotype. Additionally, we highlight a crucial nexus between calcium homeostasis (sensing, mobilization and absorption) and immune/GPCR signaling within left-sided tumors, possibly contributing to its reduced proliferative and metastatic potential. Interestingly, two genes SLC6A4 and HOXB13 show opposing regulatory trends within right and left tumors. Post-transcriptional regulation mediated by both RNA-binding proteins (e.g. NKRF (in left) and MSI2 (in right)) and miRNAs (e.g. miR-29a (in left); miR-155, miR181-d, miR-576 and miR23a (in right)) appear to exhibit side-specificity in control of their target transcripts and is pronounced in right colon tumors. Additionally, methylation results depict location-specific differences, with increased hypomethylation in open seas within left tumors, and increased hypermethylation of CpG islands within right tumors.ConclusionsDifferences in molecular mechanisms captured here highlight distinctions in tumorigenesis and progression between left and right colon tumors, which will serve as the basis for future studies, influencing the efficacies of existing and future diagnostic, prognostic and therapeutic interventions.

Project description:In the current era of precision oncology, it is widely acknowledged that CRC is a heterogeneous disease entity. Tumor location (right- or left-sided colon cancer or rectal cancer) is a crucial factor in determining disease progression as well as prognosis and influences disease management. In the last decade, numerous works have reported that the microbiome is an important element of CRC carcinogenesis, progression and therapy response. Owing to the heterogeneous nature of microbiomes, the findings of these studies were inconsistent. The majority of the studies combined colon cancer (CC) and rectal cancer (RC) samples as CRC for analysis. Furthermore, the small intestine, as the major site for immune surveillance in the gut, is understudied compared to the colon. Thus, the CRC heterogeneity puzzle is far from being solved, and more research is necessary for prospective trials that separately investigate CC and RC. Our prospective study aimed to map the colon cancer landscape using 16S rRNA amplicon sequencing in biopsy samples from the terminal ileum, healthy colon tissue, healthy rectal tissue and tumor tissue as well as in preoperative and postoperative stool samples of 41 patients. While fecal samples provide a good approximation of the average gut microbiome composition, mucosal biopsies allow for detecting subtle variations in local microbial communities. In particular, the small bowel microbiome has remained poorly characterized, mainly because of sampling difficulties. Our analysis revealed the following: (i) right- and left-sided colon cancers harbor distinct and diverse microbiomes, (ii) the tumor microbiome leads to a more consistent cancer-defined microbiome between locations and reveals a tumor microbiome-ileal microbiome association, (iii) the stool only partly reflects the microbiome landscape in patients with CC, and (iv) mechanical bowel preparation and perioperative antibiotics together with surgery result in major changes in the stool microbiome, characterized by a significant increase in the abundance of potentially pathogenic bacteria, such as Enterococcus. Collectively, our results provide new and valuable insights into the complex microbiome landscape in patients with colon cancer.

Dataset Information

Discovering the molecular differences between right- and left-sided colon cancer using machine learning methods.

Publications

Discovering the molecular differences between right- and left-sided colon cancer using machine learning methods.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets