Modeling miRNA-mRNA interactions that cause phenotypic abnormality in breast cancer patients

Metastasis-associated lung adenocarcinoma transcript 1 molecular mechanisms in gastric cancer progression

Gastric cancer (GC) remains among the most common cancers worldwide with a high mortality-to-incidence ratio. Accumulated evidence suggests that long noncoding RNAs (lncRNAs) are involved in gastric carcinogenesis. These transcripts are longer than 200 nucleotides and modulate gene expression at multiple molecular levels, inducing or inhibiting biological processes and diseases. Metastasis-associated lung adenocarcinoma transcript 1 (MALAT1) is one of the best-studied lncRNAs with comprehensive actions contributing to cancer progression. This lncRNA regulates gene expression at the transcriptional and posttranscriptional levels through interactions with microRNAs and proteins. In the present review, we discussed the molecular mechanism of MALAT1 and summarized the current knowledge of its expression in GC. Moreover, we highlighted the potential use of MALAT1 as a biomarker, including liquid biopsy.

Expression of Selected miRNAs in Normal and Cancer-Associated Fibroblasts and in BxPc3 and MIA PaCa-2 Cell Lines of Pancreatic Ductal Adenocarcinoma

Therapy for pancreatic ductal adenocarcinoma remains challenging, and the chances of a complete cure are very limited. As in other types of cancer, the expression and role of miRNAs in controlling the biological properties of this type of tumor have been extensively studied. A better insight into miRNA biology seems critical to refining diagnostics and improving their therapeutic potential. In this study, we focused on the expression of miR-21, -96, -196a, -210, and -217 in normal fibroblasts, cancer-associated fibroblasts prepared from a ductal adenocarcinoma of the pancreas, and pancreatic carcinoma cell lines. We compared these data with miRNAs in homogenates of paraffin-embedded sections from normal pancreatic tissues. In cancer-associated fibroblasts and cancer cell lines, miRNAs differed significantly from the normal tissue. In detail, miR-21 and -210 were significantly upregulated, while miR-217 was downregulated. Similar transcription profiles were earlier reported in cancer-associated fibroblasts exposed to hypoxia. However, the cells in our study were cultured under normoxic conditions. We also noted a relation to IL-6 production. In conclusion, cultured cancer-associated fibroblasts and carcinoma cells reflect miR-21 and -210 expression similarly to the cancer tissue samples harvested from the patients.

Imaging genetic association analysis of triple-negative breast cancer based on the integration of prior sample information

Triple-negative breast cancer (TNBC) is one of the more aggressive subtypes of breast cancer. The prognosis of TNBC patients remains low. Therefore, there is still a need to continue identifying novel biomarkers to improve the prognosis and treatment of TNBC patients. Research in recent years has shown that the effective use and integration of information in genomic data and image data will contribute to the prediction and prognosis of diseases. Considering that imaging genetics can deeply study the influence of microscopic genetic variation on disease phenotype, this paper proposes a sample prior information-induced multidimensional combined non-negative matrix factorization (SPID-MDJNMF) algorithm to integrate the Whole-slide image (WSI), mRNAs expression data, and miRNAs expression data. The algorithm effectively fuses high-dimensional data of three modalities through various constraints. In addition, this paper constructs an undirected graph between samples, uses an adjacency matrix to constrain the similarity, and embeds the clinical stage information of patients in the algorithm so that the algorithm can identify the co-expression patterns of samples with different labels. We performed univariate and multivariate Cox regression analysis on the mRNAs and miRNAs in the screened co-expression modules to construct a TNBC-related prognostic model. Finally, we constructed prognostic models for 2-mRNAs (IL12RB2 and CNIH2) and 2-miRNAs (miR-203a-3p and miR-148b-3p), respectively. The prognostic model can predict the survival time of TNBC patients with high accuracy. In conclusion, our proposed SPID-MDJNMF algorithm can efficiently integrate image and genomic data. Furthermore, we evaluated the prognostic value of mRNAs and miRNAs screened by the SPID-MDJNMF algorithm in TNBC, which may provide promising targets for the prognosis of TNBC patients.

Evaluation of the Relative Frequency of Epstein-Barr Virus Infection in Patients with Recurrent Breast Cancer Compared with Patients with Nonrecurrent Breast Cancer

Background

The roles of Epstein-Barr virus (EBV) in breast cancer and breast lymphoma by transfecting EBV DNA have been indicated in different studies, but few investigations have been conducted on its roles in recurrence of breast cancer. Here, we aimed to evaluate the roles of EBV in recurrent breast cancer tissue.

Materials and Methods

This is a cross-sectional retrospective study that was performed in 2020-2021 in Isfahan on patients with breast cancer. The study population consisted of 30 tissue samples from recurrent breast cancer and 30 samples from nonrecurrent breast cancer. We collected demographic data of patients including age using a checklist. Other collected data were type of cancer, stages of cancer, tumor size in greatest dimension, lymph node involvements, and presence of metastasis. Furthermore, we evaluated all of the pathology samples from both groups for the presence of DNA of EBV and compared the data of both groups.

Results

The DNA of EBV was positive in 8 patients of the relapsed group (26.6%) and 7 patients in the nonrelapsed patients (23.3%). There was no significant difference between two groups regarding positive DNA of EBV (P = 0.39). There were no significant differences between two groups of positive DNA of EBV with and without recurrent breast cancer regarding type of cancer (P = 0.63), stage of cancer (P = 0.19), tumor size in greatest dimension (P = 0.31), mean lymph node involvement (P = 0.27), number of lymph node involvement (P = 0.43), and metastasis (P = 0.69).

Conclusion

EBV might have no significant role in recurrence of breast cancer.

A review of causal discovery methods for molecular network analysis

BACKGROUND

With the increasing availability and size of multi-omics datasets, investigating the casual relationships between molecular phenotypes has become an important aspect of exploring underlying biology andgenetics. There are an increasing number of methodlogies that have been developed and applied to moleular networks to investigate these causal interactions.

METHODS

We have introduced and reviewed the available methods for building large-scale causal molecular networks that have been developed and applied in the past decade.

RESULTS

In this review we have identified and summarized the existing methods for infering causality in large-scale causal molecular networks, and discussed important factors that will need to be considered in future research in this area.

CONCLUSION

Existing methods to infering causal molecular networks have their own strengths and limitations so there is no one best approach, and it is instead down to the discretion of the researcher. This review also to discusses some of the current limitations to biological interpretation of these networks, and important factors to consider for future studies on molecular networks.

Deep Learning and Machine Learning with Grid Search to Predict Later Occurrence of Breast Cancer Metastasis Using Clinical Data

Background: It is important to be able to predict, for each individual patient, the likelihood of later metastatic occurrence, because the prediction can guide treatment plans tailored to a specific patient to prevent metastasis and to help avoid under-treatment or over-treatment. Deep neural network (DNN) learning, commonly referred to as deep learning, has become popular due to its success in image detection and prediction, but questions such as whether deep learning outperforms other machine learning methods when using non-image clinical data remain unanswered. Grid search has been introduced to deep learning hyperparameter tuning for the purpose of improving its prediction performance, but the effect of grid search on other machine learning methods are under-studied. In this research, we take the empirical approach to study the performance of deep learning and other machine learning methods when using non-image clinical data to predict the occurrence of breast cancer metastasis (BCM) 5, 10, or 15 years after the initial treatment. We developed prediction models using the deep feedforward neural network (DFNN) methods, as well as models using nine other machine learning methods, including naïve Bayes (NB), logistic regression (LR), support vector machine (SVM), LASSO, decision tree (DT), k-nearest neighbor (KNN), random forest (RF), AdaBoost (ADB), and XGBoost (XGB). We used grid search to tune hyperparameters for all methods. We then compared our feedforward deep learning models to the models trained using the nine other machine learning methods. Results: Based on the mean test AUC (Area under the ROC Curve) results, DFNN ranks 6th, 4th, and 3rd when predicting 5-year, 10-year, and 15-year BCM, respectively, out of 10 methods. The top performing methods in predicting 5-year BCM are XGB (1st), RF (2nd), and KNN (3rd). For predicting 10-year BCM, the top performers are XGB (1st), RF (2nd), and NB (3rd). Finally, for 15-year BCM, the top performers are SVM (1st), LR and LASSO (tied for 2nd), and DFNN (3rd). The ensemble methods RF and XGB outperform other methods when data are less balanced, while SVM, LR, LASSO, and DFNN outperform other methods when data are more balanced. Our statistical testing results show that at a significance level of 0.05, DFNN overall performs comparably to other machine learning methods when predicting 5-year, 10-year, and 15-year BCM. Conclusions: Our results show that deep learning with grid search overall performs at least as well as other machine learning methods when using non-image clinical data. It is interesting to note that some of the other machine learning methods, such as XGB, RF, and SVM, are very strong competitors of DFNN when incorporating grid search. It is also worth noting that the computation time required to do grid search with DFNN is much more than that required to do grid search with the other nine machine learning methods.

Assessment of Tissue Toxicity Risk in Breast Radiotherapy using Bayesian Networks

PURPOSE

The purpose of this analysis is to predict worsening post-treatment normal tissue toxicity in patients undergoing accelerated partial breast irradiation (APBI) therapy and to quantitatively identify which diagnostic, anatomical, and dosimetric features are contributing to these outcomes.

METHODS

A retrospective study of APBI treatments was performed using 32 features pertaining to various stages of the patient's treatment journey. These features were used to inform and construct a Bayesian network (BN) based on both statistical analysis of feature distributions and relative clinical importance. The target feature for prediction was defined as a measurable worsening of telangiectasia, subcutaneous tissue induration, or fibrosis when compared against the observed baseline. Parameter learning for the network was performed using data from the 299 patients included in the ACCEL trial and predictive performance was measured. Feature importance for the BN was quantified using a novel information-theoretic approach.

RESULTS

Cross validated performance of the BN for predicting toxicity was consistently higher when compared against conventional machine learning (ML) techniques. The measured BN receiver operating characteristic area under the curve was 0.960±0.013 against the best ML result of 0.942±0.021 using 5-fold cross validation with separate test data across 100 trials. The volume of the clinical target volume, gross target volume, and baseline toxicity measurements were found to have the highest feature importance and mutual dependence with normal tissue toxicity in the network, representing the strongest contribution to patient outcomes.

CONCLUSIONS

The BN outperformed conventional ML techniques in predicting tissue toxicity outcomes and provided deeper insight into which features are contributing to these outcomes. This article is protected by copyright. All rights reserved.

Comprehensive analyses of microRNA and mRNA expression in colorectal serrated lesions and colorectal cancer with a microsatellite instability phenotype

MicroRNA (miRNA) expression is dysregulated in human tumors, thereby contributing to tumorigenesis through altered expression of mRNA. Thus, identification of the relationships between miRNAs and mRNAs is important for evaluating the molecular mechanisms of tumors. In addition, elucidation of the molecular features of serrated lesions is essential in colorectal tumorigenesis. Here, we examined the relationships of miRNA and mRNA expressed in serrated lesions, including 26 sessile serrated lesions (SSLs), 12 traditional serrated adenomas (TSAs), and 11 colorectal cancers (CRCs) with a microsatellite instability (MSI) phenotype using crypt isolation. We divided the samples into the first and second cohorts for validation. Array-based expression analyses were used to evaluate miRNAs and mRNAs with opposite expression patterns in isolated tumor glands. In addition, we validated the relationships of miRNA/mRNA pairs in the second cohort using real-time polymerase chain reaction. We found that the expression of miRNA-5787 was correlated with reciprocal expression of two mRNAs, that is, SRRM2 and POLR2J3, in SSL samples. In TSA samples, two pairs of miRNAs/mRNAs showing opposite expression patterns, that is, miRNA-182-5p/ETF1 and miRNA-200b-3p/MYB, were identified. Ultimately, three pairs of miRNAs/mRNAs with opposite expression patterns, including miRNA-222-3p/SLC26A3, miRNA-6753-3p/FABP1, and miRNA-222-3p/OLFM4, were retained in CRC with an MSI phenotype. Finally, we performed transfection with an miR-222-3p mimic to confirm the expression of SLC26A3 and OLFM4; the results showed that ectopic expression of miR-222-3p moderately suppressed OLFM4 and downregulated SLC26A3 to some extent. Overall, our results provided basic insights into the evaluation of colorectal tumorigenesis of serrated lesions and CRC with an MSI phenotype.

A comprehensive scoping review of Bayesian networks in healthcare: Past, present and future

No comprehensive review of Bayesian networks (BNs) in healthcare has been published in the past, making it difficult to organize the research contributions in the present and identify challenges and neglected areas that need to be addressed in the future. This unique and novel scoping review of BNs in healthcare provides an analytical framework for comprehensively characterizing the domain and its current state. A literature search of health and health informatics literature databases using relevant keywords found 3810 articles that were reduced to 123. This was after screening out those presenting Bayesian statistics, meta-analysis or neural networks, as opposed to BNs and those describing the predictive performance of multiple machine learning algorithms, of which BNs were simply one type. Using the novel analytical framework, we show that: (1) BNs in healthcare are not used to their full potential; (2) a generic BN development process is lacking; (3) limitations exist in the way BNs in healthcare are presented in the literature, which impacts understanding, consensus towards systematic methodologies, practice and adoption; and (4) a gap exists between having an accurate BN and a useful BN that impacts clinical practice. This review highlights several neglected issues, such as restricted aims of BNs, ad hoc BN development methods, and the lack of BN adoption in practice and reveals to researchers and clinicians the need to address these problems. To map the way forward, the paper proposes future research directions and makes recommendations regarding BN development methods and adoption in practice.

Nerve growth factor receptor increases the tumor growth and metastatic potential of triple-negative breast cancer cells

Cellular heterogeneity and the lack of metastatic biomarkers limit the diagnosis of and development of therapies for metastatic triple-negative breast cancer (TNBC). Thus, development of new clinically relevant markers is urgently needed. By using RNA-seq analysis, we found that nerve growth factor receptor (NGFR) was highly expressed in metastatic lung clones of MDA-MB-231 cells. This high level of NGFR expression was necessary for TNBC cells to grow into tumor spheres under nonadhesive conditions, resist anoikis, promote primary tumor growth and increase metastasis in mice. NGFR was also expressed at a high level in a greater number of TNBC patients (45%) than non-TNBC patients (23%), enriched in higher grade tumors, and negatively correlated with the overall survival of TNBC patients. Mechanistic analysis indicated that NGFR exerted its prometastatic effects by binding with neurotrophic receptor tyrosine kinase 3 (TrkC) mainly through a ligand-independent manner, which activated the MEK-ERK1-ZEB1 and PI3K-AKT signaling pathways, increased the level of fibronectin, and decreased the expression of PUMA. Notably, we observed that NGFR expression in TrkC-positive metastatic clones reduced cellular sensitivity to anti-Trk therapy. Moreover, WNT family member 5a (WNT5A) and TrkC activated NGFR transcription in a ZEB1-dependent manner. Taken together, this study identified NGFR as a novel driver for transforming TNBC into higher grade metastatic tumors. Our findings provide the basis for the future development of NGFR as a diagnostic and prognostic marker for determining the metastatic potential of TNBC and as a therapeutic target for treating TNBC patients.

Leveraging Bayesian networks and information theory to learn risk factors for breast cancer metastasis

Background

Even though we have established a few risk factors for metastatic breast cancer (MBC) through epidemiologic studies, these risk factors have not proven to be effective in predicting an individual’s risk of developing metastasis. Therefore, identifying critical risk factors for MBC continues to be a major research imperative, and one which can lead to advances in breast cancer clinical care. The objective of this research is to leverage Bayesian Networks (BN) and information theory to identify key risk factors for breast cancer metastasis from data.

Methods

We develop the Markov Blanket and Interactive risk factor Learner (MBIL) algorithm, which learns single and interactive risk factors having a direct influence on a patient’s outcome. We evaluate the effectiveness of MBIL using simulated datasets, and compare MBIL with the BN learning algorithms Fast Greedy Search (FGS), PC algorithm (PC), and CPC algorithm (CPC). We apply MBIL to learn risk factors for 5 year breast cancer metastasis using a clinical dataset we curated. We evaluate the learned risk factors by consulting with breast cancer experts and literature. We further evaluate the effectiveness of MBIL at learning risk factors for breast cancer metastasis by comparing it to the BN learning algorithms Necessary Path Condition (NPC) and Greedy Equivalent Search (GES).

Results

The averages of the Jaccard index for the simulated datasets containing 2000 records were 0.705, 0.272, 0.228, and 0.147 for MBIL, FGS, PC, and CPC respectively. MBIL, NPC, and GES all learned that grade and lymph_nodes_positive are direct risk factors for 5 year metastasis. Only MBIL and NPC found that surgical_margins is a direct risk factor. Only NPC found that invasive is a direct risk factor. MBIL learned that HER2 and ER interact to directly affect 5 year metastasis. Neither GES nor NPC learned that HER2 and ER are direct risk factors.

Discussion

The results involving simulated datasets indicated that MBIL can learn direct risk factors substantially better than standard Bayesian network learning algorithms. An application of MBIL to a real breast cancer dataset identified both single and interactive risk factors that directly influence breast cancer metastasis, which can be investigated further.

Starvation stress attenuates the miRNA-target interaction in suppressing breast cancer cell proliferation

Background

Emerging evidence has demonstrated the limited access to metabolic substrates as an effective approach to block cancer cell growth. The mechanisms remain unclear. Our previous work has revealed that miR-221/222 plays important role in regulating breast cancer development and progression through interaction with target gene p27.

Results

Herein, we determined the miRNA-mRNA interaction in breast cancer cells under induced stress status of starvation. Starvation stimulation attenuated the miR-221/222-p27 interaction in MDA-MB-231 cells, thereby increased p27 expression and suppressed cell proliferation. Through overexpression or knockdown of miR-221/222, we found that starvation-induced stress attenuated the negative regulation of p27 expression by miR-221/222. Similar patterns for miRNA-target mRNA interaction were observed between miR-17-5p and CyclinD1, and between mR-155 and Socs1. Expression of Ago2, one of the key components of RNA-induced silencing complex (RISC), was decreased under starvation-induced stress status, which took responsibility for the impaired miRNA-target interaction since addition of exogenous Ago2 into MDA-MB-231 cells restored the miR-221/222-p27 interaction in starvation condition.

Conclusions

We demonstrated the attenuated interaction between miR-221/222 and p27 by starvation-induced stress in MDA-MB-231 breast cancer cells. The findings add a new page to the general knowledge of negative regulation of gene expression by miRNAs, also demonstrate a novel mechanism through which limited access to nutrients suppresses cancer cell proliferation. These insights provide a basis for development of novel therapeutic options for breast cancer.

Frequency of MicroRNA Response Elements Identifies Pathologically Relevant Signaling Pathways in Triple-Negative Breast Cancer

Complex interactions between mRNAs and microRNAs influence cellular functions. The mRNA-microRNA interactions also determine the post-transcriptional availability of mRNAs and unbound microRNAs. MicroRNAs binds to one or more microRNA response elements (MREs) located on the 3′UTR of mRNAs. In this study, we leveraged MREs and their frequencies in cancer and matched normal tissues to obtain insights into disease-specific interactions between mRNAs and microRNAs. We developed a bioinformatics method “ReMIx” that utilizes RNA sequencing (RNA-Seq) data to quantify MRE frequencies across the transcriptome. We applied ReMIx to triple-negative (TN) breast cancer tumor-normal adjacent pairs and identified MREs specific to TN tumors. ReMIx identified candidate mRNAs and microRNAs in the MAPK signaling cascade. Further analysis of MAPK gene regulatory networks revealed microRNA partners that influence and modulate MAPK signaling. In conclusion, we demonstrate a novel method of using MREs in the identification of functionally relevant mRNA-microRNA interactions in TN breast cancer.

•

Bioinformatics method ReMIx identify differential microRNA response rlements (MRE)

•

Tumor-specific MREs frequency observed in triple-negative breast cancer (TNBC)

•

MRE analysis identify MAPK signaling genes as therapeutic target for TNBC

•

MREs frequency can be used to identify pathologically relevant pathways

Data Mining and Expression Analysis of Differential lncRNA ADAMTS9-AS1 in Prostate Cancer

Long noncoding RNAs (lncRNAs) play important roles in the regulation of gene expression by acting as competing endogenous RNAs (ceRNAs). However, the roles of lncRNA-associated ceRNAs in oncogenesis are not fully understood. The present study aims to determine whether a ceRNA network can serve as a prognostic marker in human prostate cancer (PCa). In order to identify a ceRNA network and the key lncRNAs in PCa, we constructed a differentially expressed lncRNAs (DELs)-differentially expressed miRNAs (DEMis)-differentially expressed mRNAs (DEMs) regulatory network based on the ceRNA theory using data from the Cancer Genome Atlas (TCGA). We found that the DELs-DEMis-DEMs network was composed of 27 DELs nodes, seven DEMis nodes, and three DEMs nodes. The 27 DELs were further analyzed with several public databases to provide meaningful information for understanding the functional roles of lncRNAs in regulatory networks in PCa. We selected ADAMTS9-AS1 to determine its role in PCa and found that ADAMTS9-AS1 significantly influences tumor cell growth and proliferation, suggesting that it plays a tumor suppressive role. In addition, ADAMTS9-AS1 functioned as ceRNA, effectively becoming a sponge for hsa-mir-96 and modulating the expression of PRDM16. These results suggest that ceRNAs could accelerate biomarker discovery and therapeutic strategies for PCa.

The fundamental role of miR-10b in metastatic cancer

Small, non-coding strands of RNA have been identified as a significant player in the pathology of cancer. One of the first miRNAs to be shown as having aberrant expression in cancer was miR-10b. Since the inaugural study on miR-10b, its role as a metastasis promoting factor has been extensively validated. To date, more than 100 studies have been completed on miR-10b and metastasis across 18 cancer types. This immense set of information holds possibilities for novel methods to improve the lives of many. This review outlines what is currently understood of miR-10b's clinical significance, its molecular regulation, and the possible diagnostic and therapeutic methods leveraging miR-10b as a fundamental target in metastatic cancer. Such methods would move us closer to developing a truly individualized therapeutic strategy against cancer and will likely provide unique information about cancer staging, disease outcome, metastatic potential, and ultimately survival.

Co-Expression Network Analysis Identifies miRNA⁻mRNA Networks Potentially Regulating Milk Traits and Blood Metabolites

MicroRNAs (miRNA) regulate mRNA networks to coordinate cellular functions. In this study, we constructed gene co-expression networks to detect miRNA modules (clusters of miRNAs with similar expression patterns) and miRNA–mRNA pairs associated with blood (triacylglyceride and nonesterified fatty acids) and milk (milk yield, fat, protein, and lactose) components and milk fatty acid traits following dietary supplementation of cows’ diets with 5% linseed oil (LSO) (n = 6 cows) or 5% safflower oil (SFO) (n = 6 cows) for 28 days. Using miRNA transcriptome data from mammary tissues of cows for co-expression network analysis, we identified three consensus modules: blue, brown, and turquoise, composed of 70, 34, and 86 miRNA members, respectively. The hub miRNAs (miRNAs with the most connections with other miRNAs) were miR-30d, miR-484 and miR-16b for blue, brown, and turquoise modules, respectively. Cell cycle arrest, and p53 signaling and transforming growth factor–beta (TGF-β) signaling pathways were the common gene ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathways enriched for target genes of the three modules. Protein percent (p = 0.03) correlated with the turquoise module in LSO treatment while protein yield (p = 0.003) and milk yield (p = 7 × 10⁻⁰⁴) correlated with the turquoise model, protein and milk yields and lactose percent (p < 0.05) correlated with the blue module and fat percent (p = 0.04) correlated with the brown module in SFO treatment. Several fatty acids correlated (p < 0.05) with the blue (CLA:9,11) and brown (C4:0, C12:0, C22:0, C18:1n9c and CLA:10,12) modules in LSO treatment and with the turquoise (C14:0, C18:3n3 and CLA:9,11), blue (C14:0 and C23:0) and brown (C6:0, C16:0, C22:0, C22:6n3 and CLA:10,12) modules in SFO treatment. Correlation of miRNA and mRNA data from the same animals identified the following miRNA–mRNA pairs: miR-183/RHBDD2 (p = 0.003), miR-484/EIF1AD (p = 0.011) and miR-130a/SBSPON (p = 0.004) with lowest p-values for the blue, brown, and turquoise modules, respectively. Milk yield, protein yield, and protein percentage correlated (p < 0.05) with 28, 31 and 5 miRNA–mRNA pairs, respectively. Our results suggest that, the blue, brown, and turquoise modules miRNAs, hub miRNAs, miRNA–mRNA networks, cell cycle arrest GO term, p53 signaling and TGF-β signaling pathways have considerable influence on milk and blood phenotypes following dietary supplementation of dairy cows’ diets with 5% LSO or 5% SFO.

Machine Learning and Radiogenomics: Lessons Learned and Future Directions

Due to the rapid increase in the availability of patient data, there is significant interest in precision medicine that could facilitate the development of a personalized treatment plan for each patient on an individual basis. Radiation oncology is particularly suited for predictive machine learning (ML) models due to the enormous amount of diagnostic data used as input and therapeutic data generated as output. An emerging field in precision radiation oncology that can take advantage of ML approaches is radiogenomics, which is the study of the impact of genomic variations on the sensitivity of normal and tumor tissue to radiation. Currently, patients undergoing radiotherapy are treated using uniform dose constraints specific to the tumor and surrounding normal tissues. This is suboptimal in many ways. First, the dose that can be delivered to the target volume may be insufficient for control but is constrained by the surrounding normal tissue, as dose escalation can lead to significant morbidity and rare. Second, two patients with nearly identical dose distributions can have substantially different acute and late toxicities, resulting in lengthy treatment breaks and suboptimal control, or chronic morbidities leading to poor quality of life. Despite significant advances in radiogenomics, the magnitude of the genetic contribution to radiation response far exceeds our current understanding of individual risk variants. In the field of genomics, ML methods are being used to extract harder-to-detect knowledge, but these methods have yet to fully penetrate radiogenomics. Hence, the goal of this publication is to provide an overview of ML as it applies to radiogenomics. We begin with a brief history of radiogenomics and its relationship to precision medicine. We then introduce ML and compare it to statistical hypothesis testing to reflect on shared lessons and to avoid common pitfalls. Current ML approaches to genome-wide association studies are examined. The application of ML specifically to radiogenomics is next presented. We end with important lessons for the proper integration of ML into radiogenomics.

MiR-192-5p suppresses the growth of bladder cancer cells via targeting Yin Yang 1

Bladder cancer has been identified as one of the most malignant cancers with high incidence and mortality. The underlying mechanisms by which regulate the tumorigenesis of bladder cancer deserve further investigation. Here, we found that miR-192-5p was downregulated in human bladder cancer cell lines and tissues. Overexpression of miR-192-5p significantly inhibited the growth of bladder cancer cells, while depletion of miR-192-5p exerted opposite effect. Bioinformatics analysis and molecular mechanism study identified that miR-192-5p targeted the transcription factor Yin Yang 1 (YY1) and decreased the expression level of YY1. Highly expressed YY1 attenuated the potential tumor suppressive function of miR-192-5p. The expression of miR-192-5p was negatively correlated with that of YY1 in bladder cancer tissues. These results indicated that miR-192-5p might serve as a promising target in bladder cancer diagnosis and therapy.