Special Issue "Application of Bioinformatics in Cancers"

A special issue of Cancers (ISSN 2072-6694).

Deadline for manuscript submissions: 31 March 2019

Special Issue Editor

Guest Editor
Dr. J. Chad Brenner

Assistant Professor, Department of Otolaryngology-Head and Neck Surgery
Director, Michigan Otolaryngology and Translational Oncology Laboratory
University of Michigan Health Systems, Ann Arbor, MI 48109, USA
Website | E-Mail
Interests: functional genomic, proteomic and bioinformatics approaches in cancer, sequencing the exomes and transcriptomes of head and neck cancer; drug sensitivities

Special Issue Information

Dear Colleagues,

Bioinformatics applications in cancer have rapidly evolved over the past several years. Ever since its initial implementation, next generation sequencing has altered our understanding of cancer biology, and the approaches to analyze more and more complex datasets have also become increasingly complex.  Routine bioinformatics pipelines now range from those that rapidly detect and predict functional impact of molecular alterations to those that quantify changes to the tumor microenvironment. For example, several tools that analyze tumor-immune interactions have been successfully developed to assess tumor infiltrating lymphocyte content, microsatellite instability, total mutational burden and neoantigen presentation. Further complexity of integrated omics-based analysis is also now coupled with the emergence of modern machine learning and network-based approaches to analyze large datasets in the context of publicly available resources, such as the cancer genome atlas.

While much of the focus has so far been on annotating molecular alterations as well as infiltrating cell types or cell states in ideal sequencing conditions, alternative and application-specific approaches are now emerging that improve on a wide variety of established analysis techniques. These include techniques that range from improved quantification of copy number and gene expression from formalin fixed tissues as well as applications that require high sensitivity such as the quantification of tumor mutations from liquid biopsies (circulating cell free DNA).  Further novel applications attempt to improve the ability to analyze the distribution and molecular impact of complicated genetic features such as repetitive or transposable endogenous elements (e.g., LINE-1) as well as exogenous genetic elements (e.g., human papilloma virus).

As we develop a better understanding of the limitations of these new informatics approaches, we can ultimately hope to apply these techniques to existing datasets and build well-annotated databases of easily accessible information that can be leveraged in multi-variable analysis pipelines. Similar to the success of SIGdb and cBioPortal, this should help yield new diagnostic and prognostic/predictive biomarkers for standard interventional modalities as well as emerging areas like immuno-oncology, and areas of unmet clinical need. This Special Issue will highlight the current state of the art in bioinformatics applications in cancer biology, and infer future prospects for improving informatics applications through artificial intelligence and machine learning approaches.

Dr. J. Chad Brenner
Guest Editor

Manuscript Submission Information

Manuscripts should be submitted online at www.mdpi.com by registering and logging in to this website. Once you are registered, click here to go to the submission form. Manuscripts can be submitted until the deadline. All papers will be peer-reviewed. Accepted papers will be published continuously in the journal (as soon as accepted) and will be listed together on the special issue website. Research articles, review articles as well as short communications are invited. For planned papers, a title and short abstract (about 100 words) can be sent to the Editorial Office for announcement on this website.

Submitted manuscripts should not have been published previously, nor be under consideration for publication elsewhere (except conference proceedings papers). All manuscripts are thoroughly refereed through a single-blind peer-review process. A guide for authors and other relevant information for submission of manuscripts is available on the Instructions for Authors page. Cancers is an international peer-reviewed open access monthly journal published by MDPI.

Please visit the Instructions for Authors page before submitting a manuscript. The Article Processing Charge (APC) for publication in this open access journal is 1800 CHF (Swiss Francs). Submitted papers should be well formatted and use good English. Authors may use MDPI's English editing service prior to publication or during author revisions.

Keywords

  • Bioinformatics
  • Machine learning
  • Artificial Intelligence
  • Network Analysis
  • Single cell sequencing
  • Circulating tumor DNA (ctDNA)
  • Neoantigen Prediction
  • Precision medicine
  • Computational Immunology

Published Papers (16 papers)

View options order results:
result details:
Displaying articles 1-16
Export citation of selected articles as:

Research

Jump to: Review, Other

Open AccessArticle Validation of miRNAs as Breast Cancer Biomarkers with a Machine Learning Approach
Cancers 2019, 11(3), 431; https://doi.org/10.3390/cancers11030431 (registering DOI)
Received: 28 February 2019 / Revised: 19 March 2019 / Accepted: 20 March 2019 / Published: 26 March 2019
PDF Full-text (313 KB) | HTML Full-text | XML Full-text
Abstract
Certain small noncoding microRNAs (miRNAs) are differentially expressed in normal tissues and cancers, which makes them great candidates for biomarkers for cancer. Previously, a selected subset of miRNAs has been experimentally verified to be linked to breast cancer. In this paper, we validated [...] Read more.
Certain small noncoding microRNAs (miRNAs) are differentially expressed in normal tissues and cancers, which makes them great candidates for biomarkers for cancer. Previously, a selected subset of miRNAs has been experimentally verified to be linked to breast cancer. In this paper, we validated the importance of these miRNAs using a machine learning approach on miRNA expression data. We performed feature selection, using Information Gain (IG), Chi-Squared (CHI2) and Least Absolute Shrinkage and Selection Operation (LASSO), on the set of these relevant miRNAs to rank them by importance. We then performed cancer classification using these miRNAs as features using Random Forest (RF) and Support Vector Machine (SVM) classifiers. Our results demonstrated that the miRNAs ranked higher by our analysis had higher classifier performance. Performance becomes lower as the rank of the miRNA decreases, confirming that these miRNAs had different degrees of importance as biomarkers. Furthermore, we discovered that using a minimum of three miRNAs as biomarkers for breast cancers can be as effective as using the entire set of 1800 miRNAs. This work suggests that machine learning is a useful tool for functional studies of miRNAs for cancer detection and diagnosis. Full article
(This article belongs to the Special Issue Application of Bioinformatics in Cancers)
Figures

Figure 1

Open AccessArticle A Systematic Pan-Cancer Analysis of Genetic Heterogeneity Reveals Associations with Epigenetic Modifiers
Cancers 2019, 11(3), 391; https://doi.org/10.3390/cancers11030391
Received: 12 February 2019 / Revised: 9 March 2019 / Accepted: 17 March 2019 / Published: 20 March 2019
PDF Full-text (2962 KB) | HTML Full-text | XML Full-text | Supplementary Files
Abstract
Intratumor genetic heterogeneity (ITH) is the main obstacle to effective cancer treatment and a major mechanism of drug resistance. It results from the continuous evolution of different clones of a tumor over time. However, the molecular features underlying the emergence of genetically-distinct subclonal [...] Read more.
Intratumor genetic heterogeneity (ITH) is the main obstacle to effective cancer treatment and a major mechanism of drug resistance. It results from the continuous evolution of different clones of a tumor over time. However, the molecular features underlying the emergence of genetically-distinct subclonal cell populations remain elusive. Here, we conducted an exhaustive characterization of ITH across 2807 tumor samples from 16 cancer types. Integration of ITH scores and somatic variants detected in each tumor sample revealed that mutations in epigenetic modifier genes are associated with higher ITH levels. In particular, genes that regulate genome-wide histone and DNA methylation emerged as being determinant of high ITH. Indeed, the knockout of histone methyltransferase SETD2 or DNA methyltransferase DNMT3A using the CRISPR/Cas9 system on cancer cells led to significant expansion of genetically-distinct clones and culminated in highly heterogeneous cell populations. The ITH scores observed in knockout cells recapitulated the heterogeneity levels observed in patient tumor samples and correlated with a better mitochondrial bioenergetic performance under stress conditions. Our work provides new insights into tumor development, and discloses new drivers of ITH, which may be useful as either predictive biomarkers or therapeutic targets to improve cancer treatment. Full article
(This article belongs to the Special Issue Application of Bioinformatics in Cancers)
Figures

Graphical abstract

Open AccessArticle Examination of Independent Prognostic Power of Gene Expressions and Histopathological Imaging Features in Cancer
Cancers 2019, 11(3), 361; https://doi.org/10.3390/cancers11030361
Received: 12 February 2019 / Revised: 4 March 2019 / Accepted: 10 March 2019 / Published: 13 March 2019
PDF Full-text (1775 KB) | HTML Full-text | XML Full-text | Supplementary Files
Abstract
Cancer prognosis is of essential interest, and extensive research has been conducted searching for biomarkers with prognostic power. Recent studies have shown that both omics profiles and histopathological imaging features have prognostic power. There are also studies exploring integrating the two types of [...] Read more.
Cancer prognosis is of essential interest, and extensive research has been conducted searching for biomarkers with prognostic power. Recent studies have shown that both omics profiles and histopathological imaging features have prognostic power. There are also studies exploring integrating the two types of measurements for prognosis modeling. However, there is a lack of study rigorously examining whether omics measurements have independent prognostic power conditional on histopathological imaging features, and vice versa. In this article, we adopt a rigorous statistical testing framework and test whether an individual gene expression measurement can improve prognosis modeling conditional on high-dimensional imaging features, and a parallel analysis is conducted reversing the roles of gene expressions and imaging features. In the analysis of The Cancer Genome Atlas (TCGA) lung adenocarcinoma and liver hepatocellular carcinoma data, it is found that multiple individual genes, conditional on imaging features, can lead to significant improvement in prognosis modeling; however, individual imaging features, conditional on gene expressions, only offer limited prognostic power. Being among the first to examine the independent prognostic power, this study may assist better understanding the “connectedness” between omics profiles and histopathological imaging features and provide important insights for data integration in cancer modeling. Full article
(This article belongs to the Special Issue Application of Bioinformatics in Cancers)
Figures

Figure 1

Open AccessArticle Observed Survival Interval: A Supplement to TCGA Pan-Cancer Clinical Data Resource
Cancers 2019, 11(3), 280; https://doi.org/10.3390/cancers11030280
Received: 24 January 2019 / Revised: 16 February 2019 / Accepted: 22 February 2019 / Published: 26 February 2019
PDF Full-text (2634 KB) | HTML Full-text | XML Full-text | Supplementary Files
Abstract
To drive high-quality omics translational research using The Cancer Genome Atlas (TCGA) data, a TCGA Pan-Cancer Clinical Data Resource was proposed. However, there is an out-of-step issue between clinical outcomes and the omics data of TCGA for skin cutaneous melanoma (SKCM), due to [...] Read more.
To drive high-quality omics translational research using The Cancer Genome Atlas (TCGA) data, a TCGA Pan-Cancer Clinical Data Resource was proposed. However, there is an out-of-step issue between clinical outcomes and the omics data of TCGA for skin cutaneous melanoma (SKCM), due to the majority of metastatic samples. In clinical cases, the survival time started from the initial SKCM diagnosis, while the omics data were characterized at TCGA sampling. This study aimed to address this issue by proposing an observed survival interval (OBS), which was defined as the time interval from TCGA sampling to patient death or last follow-up. We compared the OBS with the usual recommended overall survival (OS) by associating them with both clinical data and microRNA sequencing data of TCGA-SKCM. We found that the OS of primary SKCM was significantly shorter than that of metastatic SKCM, while the opposite happened if OBS was compared. OS was associated with the pathological stage of both primary and metastatic SKCM, while OBS was associated with the pathological stage of primary SKCM but not that of metastatic SKCM. Five previously cross-validated survival-associated microRNAs were found to be associated with the OBS rather than OS in metastatic SKCM. Thus, the OBS was more appropriate for associating microRNA-omics data of TCGA-SKCM than OS, and it is a timely supplement to TCGA Pan-Cancer Clinical Data Resource. Full article
(This article belongs to the Special Issue Application of Bioinformatics in Cancers)
Figures

Figure 1

Open AccessArticle Developing a Prognostic Gene Panel of Epithelial Ovarian Cancer Patients by a Machine Learning Model
Cancers 2019, 11(2), 270; https://doi.org/10.3390/cancers11020270
Received: 11 January 2019 / Revised: 18 February 2019 / Accepted: 22 February 2019 / Published: 25 February 2019
PDF Full-text (1051 KB) | HTML Full-text | XML Full-text | Supplementary Files
Abstract
Epithelial ovarian cancer patients usually relapse after primary management. We utilized the support vector machine algorithm to develop a model for the chemo-response using the Cancer Cell Line Encyclopedia (CCLE) and validated the model in The Cancer Genome Atlas (TCGA) and the GSE9891 [...] Read more.
Epithelial ovarian cancer patients usually relapse after primary management. We utilized the support vector machine algorithm to develop a model for the chemo-response using the Cancer Cell Line Encyclopedia (CCLE) and validated the model in The Cancer Genome Atlas (TCGA) and the GSE9891 dataset. Finally, we evaluated the feasibility of the model using ovarian cancer patients from our institute. The 10-gene predictive model demonstrated that the high response group had a longer recurrence-free survival (RFS) (log-rank test, p = 0.015 for TCGA, p = 0.013 for GSE9891 and p = 0.039 for NTUH) and overall survival (OS) (log-rank test, p = 0.002 for TCGA and p = 0.016 for NTUH). In a multivariate Cox hazard regression model, the predictive model (HR: 0.644, 95% CI: 0.436–0.952, p = 0.027) and residual tumor size < 1 cm (HR: 0.312, 95% CI: 0.170–0.573, p < 0.001) were significant factors for recurrence. The predictive model (HR: 0.511, 95% CI: 0.334–0.783, p = 0.002) and residual tumor size < 1 cm (HR: 0.252, 95% CI: 0.128–0.496, p < 0.001) were still significant factors for death. In conclusion, the patients of high response group stratified by the model had good response and favourable prognosis, whereas for the patients of medium to low response groups, introduction of other drugs or clinical trials might be beneficial. Full article
(This article belongs to the Special Issue Application of Bioinformatics in Cancers)
Figures

Figure 1

Open AccessArticle Reverse Engineering Cancer: Inferring Transcriptional Gene Signatures from Copy Number Aberrations with ICAro
Cancers 2019, 11(2), 256; https://doi.org/10.3390/cancers11020256
Received: 28 December 2018 / Revised: 7 February 2019 / Accepted: 13 February 2019 / Published: 22 February 2019
PDF Full-text (1372 KB) | HTML Full-text | XML Full-text | Supplementary Files
Abstract
The characterization of a gene product function is a process that involves multiple laboratory techniques in order to silence the gene itself and to understand the resulting cellular phenotype via several omics profiling. When it comes to tumor cells, usually the translation process [...] Read more.
The characterization of a gene product function is a process that involves multiple laboratory techniques in order to silence the gene itself and to understand the resulting cellular phenotype via several omics profiling. When it comes to tumor cells, usually the translation process from in vitro characterization results to human validation is a difficult journey. Here, we present a simple algorithm to extract mRNA signatures from cancer datasets, where a particular gene has been deleted at the genomic level, ICAro. The process is implemented as a two-step workflow. The first one employs several filters in order to select the two patient subsets: the inactivated one, where the target gene is deleted, and the control one, where large genomic rearrangements should be absent. The second step performs a signature extraction via a Differential Expression analysis and a complementary Random Forest approach to provide an additional gene ranking in terms of information loss. We benchmarked the system robustness on a panel of genes frequently deleted in cancers, where we validated the downregulation of target genes and found a correlation with signatures extracted with the L1000 tool, outperforming random sampling for two out of six L1000 classes. Furthermore, we present a use case correlation with a published transcriptomic experiment. In conclusion, deciphering the complex interactions of the tumor environment is a challenge that requires the integration of several experimental techniques in order to create reproducible results. We implemented a tool which could be of use when trying to find mRNA signatures related to a gene loss event to better understand its function or for a gene-loss associated biomarker research. Full article
(This article belongs to the Special Issue Application of Bioinformatics in Cancers)
Figures

Figure 1

Open AccessArticle Nucleotide Weight Matrices Reveal Ubiquitous Mutational Footprints of AID/APOBEC Deaminases in Human Cancer Genomes
Cancers 2019, 11(2), 211; https://doi.org/10.3390/cancers11020211
Received: 11 January 2019 / Revised: 30 January 2019 / Accepted: 30 January 2019 / Published: 12 February 2019
PDF Full-text (1385 KB) | HTML Full-text | XML Full-text | Supplementary Files
Abstract
Cancer genomes accumulate nucleotide sequence variations that number in the tens of thousands per genome. A prominent fraction of these mutations is thought to arise as a consequence of the off-target activity of DNA/RNA editing cytosine deaminases. These enzymes, collectively called activation induced [...] Read more.
Cancer genomes accumulate nucleotide sequence variations that number in the tens of thousands per genome. A prominent fraction of these mutations is thought to arise as a consequence of the off-target activity of DNA/RNA editing cytosine deaminases. These enzymes, collectively called activation induced deaminase (AID)/APOBECs, deaminate cytosines located within defined DNA sequence contexts. The resulting changes of the original C:G pair in these contexts (mutational signatures) provide indirect evidence for the participation of specific cytosine deaminases in a given cancer type. The conventional method used for the analysis of mutable motifs is the consensus approach. Here, for the first time, we have adopted the frequently used weight matrix (sequence profile) approach for the analysis of mutagenesis and provide evidence for this method being a more precise descriptor of mutations than the sequence consensus approach. We confirm that while mutational footprints of APOBEC1, APOBEC3A, APOBEC3B, and APOBEC3G are prominent in many cancers, mutable motifs characteristic of the action of the humoral immune response somatic hypermutation enzyme, AID, are the most widespread feature of somatic mutation spectra attributable to deaminases in cancer genomes. Overall, the weight matrix approach reveals that somatic mutations are significantly associated with at least one AID/APOBEC mutable motif in all studied cancers. Full article
(This article belongs to the Special Issue Application of Bioinformatics in Cancers)
Figures

Figure 1

Open AccessArticle Potential Applications of DNA, RNA and Protein Biomarkers in Diagnosis, Therapy and Prognosis for Colorectal Cancer: A Study from Databases to AI-Assisted Verification
Cancers 2019, 11(2), 172; https://doi.org/10.3390/cancers11020172
Received: 11 December 2018 / Revised: 21 January 2019 / Accepted: 29 January 2019 / Published: 1 February 2019
PDF Full-text (6699 KB) | HTML Full-text | XML Full-text
Abstract
In order to find out the most valuable biomarkers and pathways for diagnosis, therapy and prognosis in colorectal cancer (CRC) we have collected the published CRC biomarkers and established a CRC biomarker database (CBD: http://sysbio.suda.edu.cn/CBD/index.html). In this study, we analysed the single and [...] Read more.
In order to find out the most valuable biomarkers and pathways for diagnosis, therapy and prognosis in colorectal cancer (CRC) we have collected the published CRC biomarkers and established a CRC biomarker database (CBD: http://sysbio.suda.edu.cn/CBD/index.html). In this study, we analysed the single and multiple DNA, RNA and protein biomarkers as well as their positions in cancer related pathways and protein-protein interaction (PPI) networks to describe their potential applications in diagnosis, therapy and prognosis. CRC biomarkers were collected from the CBD. The RNA and protein biomarkers were matched to their corresponding DNAs by the miRDB database and the PubMed Gene database, respectively. The PPI networks were used to investigate the relationships between protein biomarkers and further detect the multiple biomarkers. The Kyoto Encyclopaedia of Genes and Genomes (KEGG) pathway enrichment analysis and Gene Ontology (GO) annotation were used to analyse biological functions of the biomarkers. AI classification techniques were utilized to further verify the significances of the multiple biomarkers in diagnosis and prognosis for CRC. We showed that a large number of the DNA, RNA and protein biomarkers were associated with the diagnosis, therapy and prognosis in various degrees in the CRC biomarker networks. The CRC biomarkers were closely related to the CRC initiation and progression. Moreover, the biomarkers played critical roles in cellular proliferation, apoptosis and angiogenesis and they were involved in Ras, p53 and PI3K pathways. There were overlaps among the DNA, RNA and protein biomarkers. AI classification verifications showed that the combined multiple protein biomarkers played important roles to accurate early diagnosis and predict outcome for CRC. There were several single and multiple CRC protein biomarkers which were associated with diagnosis, therapy and prognosis in CRC. Further, AI-assisted analysis revealed that multiple biomarkers had potential applications for diagnosis and prognosis in CRC. Full article
(This article belongs to the Special Issue Application of Bioinformatics in Cancers)
Figures

Figure 1

Open AccessArticle Prognostic Biomarkers in Pancreatic Cancer: Avoiding Errata When Using the TCGA Dataset
Cancers 2019, 11(1), 126; https://doi.org/10.3390/cancers11010126
Received: 4 December 2018 / Revised: 4 January 2019 / Accepted: 16 January 2019 / Published: 21 January 2019
PDF Full-text (1578 KB) | HTML Full-text | XML Full-text | Supplementary Files
Abstract
Data from the Cancer Genome Atlas (TCGA) are now easily accessible through web-based platforms with tools to assess the prognostic value of molecular alterations. Pancreatic tumors have heterogeneous biology and aggressiveness ranging from the deadly adenocarcinoma (PDAC) to the better prognosis, neuroendocrine tumors. [...] Read more.
Data from the Cancer Genome Atlas (TCGA) are now easily accessible through web-based platforms with tools to assess the prognostic value of molecular alterations. Pancreatic tumors have heterogeneous biology and aggressiveness ranging from the deadly adenocarcinoma (PDAC) to the better prognosis, neuroendocrine tumors. We assessed the availability of the pancreatic cancer TCGA data (TCGA_PAAD) from several repositories and investigated the nature of each sample and how non-PDAC samples impact prognostic biomarker studies. While the clinical and genomic data (n = 185) were fairly consistent across all repositories, RNAseq profiles varied from 176 to 185. As a result, 35 RNAseq profiles (18.9%) corresponded to a normal, inflamed pancreas or non-PDAC neoplasms. This information was difficult to obtain. By considering gene expression data as continuous values, the expression of the 5312 and 4221 genes were significantly associated with the progression-free and overall survival respectively. Considering the cohort was not curated, only 4 and 14, respectively, had prognostic value in the PDAC-only cohort. Similarly, mutations in key genes or well-described miRNA lost their prognostic significance in the PDAC-only cohort. Therefore, we propose a web-based application to assess biomarkers in the curated TCGA_PAAD dataset. In conclusion, TCGA_PAAD curation is critical to avoid important biological and clinical biases from non-PDAC samples. Full article
(This article belongs to the Special Issue Application of Bioinformatics in Cancers)
Figures

Figure 1

Open AccessArticle The Potential Mechanism of Bufadienolide-Like Chemicals on Breast Cancer via Bioinformatics Analysis
Received: 28 November 2018 / Revised: 20 December 2018 / Accepted: 8 January 2019 / Published: 14 January 2019
PDF Full-text (3315 KB) | HTML Full-text | XML Full-text | Supplementary Files
Abstract
Bufadienolide-like chemicals are mostly composed of the active ingredient of Chansu and they have anti-inflammatory, tumor-suppressing, and anti-pain activities; however, their mechanism is unclear. This work used bioinformatics analysis to study this mechanism via gene expression profiles of bufadienolide-like chemicals: (1) Differentially expressed [...] Read more.
Bufadienolide-like chemicals are mostly composed of the active ingredient of Chansu and they have anti-inflammatory, tumor-suppressing, and anti-pain activities; however, their mechanism is unclear. This work used bioinformatics analysis to study this mechanism via gene expression profiles of bufadienolide-like chemicals: (1) Differentially expressed gene identification combined with gene set variation analysis, (2) similar small -molecule detection, (3) tissue-specific co-expression network construction, (4) differentially regulated sub-networks related to breast cancer phenome, (5) differentially regulated sub-networks with potential cardiotoxicity, and (6) hub gene selection and their relation to survival probability. The results indicated that bufadienolide-like chemicals usually had the same target as valproic acid and estradiol, etc. They could disturb the pathways in RNA splicing, the apoptotic process, cell migration, extracellular matrix organization, adherens junction organization, synaptic transmission, Wnt signaling, AK-STAT signaling, BMP signaling pathway, and protein folding. We also investigated the potential cardiotoxicity and found a dysregulated subnetwork related to membrane depolarization during action potential, retinoic acid receptor binding, GABA receptor binding, positive regulation of nuclear division, negative regulation of viral genome replication, and negative regulation of the viral life cycle. These may play important roles in the cardiotoxicity of bufadienolide-like chemicals. The results may highlight the potential anticancer mechanism and cardiotoxicity of Chansu, and could also explain the ability of bufadienolide-like chemicals to be used as hormones and anticancer and vasoprotectives agents. Full article
(This article belongs to the Special Issue Application of Bioinformatics in Cancers)
Figures

Figure 1

Open AccessArticle Personalized Prediction of Acquired Resistance to EGFR-Targeted Inhibitors Using a Pathway-Based Machine Learning Approach
Received: 7 December 2018 / Revised: 22 December 2018 / Accepted: 26 December 2018 / Published: 4 January 2019
PDF Full-text (2011 KB) | HTML Full-text | XML Full-text | Supplementary Files
Abstract
Epidermal growth factor receptor (EGFR) inhibitors have benefitted cancer patients worldwide, but resistance inevitably develops over time, resulting in treatment failures. An accurate prediction model for acquired resistance (AR) to EGFR inhibitors is critical for early diagnosis and according intervention, but is not [...] Read more.
Epidermal growth factor receptor (EGFR) inhibitors have benefitted cancer patients worldwide, but resistance inevitably develops over time, resulting in treatment failures. An accurate prediction model for acquired resistance (AR) to EGFR inhibitors is critical for early diagnosis and according intervention, but is not yet available due to personal variations and the complex mechanisms of AR. Here, we have developed a novel pipeline to build a meta-analysis-based, multivariate model for personalized pathways in AR to EGFR inhibitors, using sophisticated machine learning algorithms. Surprisingly, the model achieved excellent predictive performance, with a cross-study validation area under curve (AUC) of over 0.9, and generalization performance on independent cohorts of samples, with a perfect AUC score of 1. Furthermore, the model showed excellent transferability across different cancer cell lines and EGFR inhibitors, including gefitinib, erlotinib, afatinib, and cetuximab. In conclusion, our model achieved high predictive accuracy through robust cross study validation, and enabled individualized prediction on newly introduced data. We also discovered common pathway alteration signatures for AR to EGFR inhibitors, which can provide directions for other follow-up studies. Full article
(This article belongs to the Special Issue Application of Bioinformatics in Cancers)
Figures

Figure 1

Open AccessArticle Pathological and Molecular Characteristics of Colorectal Cancer with Brain Metastases
Cancers 2018, 10(12), 504; https://doi.org/10.3390/cancers10120504
Received: 17 November 2018 / Revised: 30 November 2018 / Accepted: 5 December 2018 / Published: 10 December 2018
PDF Full-text (391 KB) | HTML Full-text | XML Full-text
Abstract
Background: Colorectal cancers (CRC) with brain metastases (BM) are scarcely described. The main objective of this study was to determine the molecular profile of CRC with BM. Methods: We included 82 CRC patients with BM. KRAS, NRAS, BRAF and mismatch repair [...] Read more.
Background: Colorectal cancers (CRC) with brain metastases (BM) are scarcely described. The main objective of this study was to determine the molecular profile of CRC with BM. Methods: We included 82 CRC patients with BM. KRAS, NRAS, BRAF and mismatch repair (MMR) status were investigated on primary tumors (n = 82) and BM (n = 38). ALK, ROS1, cMET, HER-2, PD-1, PD-L1, CD3 and CD8 status were evaluated by immunohistochemistry, and when recommended, by fluorescence in situ hybridization. Results: In primary tumors, KRAS, NRAS and BRAF mutations were observed in 56%, 6%, and 6% of cases, respectively. No ROS1, ALK and cMET rearrangement was detected. Only one tumor presented HER-2 amplification. Molecular profiles were mostly concordant between BM and paired primary tumors, except for 9% of discordances for RAS mutation. CD3, CD8, PD-1 and PD-L1 expressions presented some discordance between primary tumors and BM. In multivariate analysis, multiple BM, lung metastases and PD-L1+ tumor were predictive of poor overall survival. Conclusions: CRCs with BM are associated with high frequency of RAS mutations and significant discordance for RAS mutational status between BM and paired primary tumors. Multiple BM, lung metastases and PD-L1+ have been identified as prognostic factors and can guide therapeutic decisions for CRC patients with BM. Full article
(This article belongs to the Special Issue Application of Bioinformatics in Cancers)
Figures

Figure 1

Open AccessArticle Network Pharmacology to Unveil the Biological Basis of Health-Strengthening Herbal Medicine in Cancer Treatment
Cancers 2018, 10(11), 461; https://doi.org/10.3390/cancers10110461
Received: 12 September 2018 / Revised: 5 November 2018 / Accepted: 16 November 2018 / Published: 21 November 2018
Cited by 1 | PDF Full-text (26745 KB) | HTML Full-text | XML Full-text | Supplementary Files
Abstract
Health-strengthening (Fu-Zheng) herbs is a representative type of traditional Chinese medicine (TCM) widely used for cancer treatment in China, which is in contrast to pathogen eliminating (Qu-Xie) herbs. However, the commonness in the biological basis of health-strengthening herbs remains [...] Read more.
Health-strengthening (Fu-Zheng) herbs is a representative type of traditional Chinese medicine (TCM) widely used for cancer treatment in China, which is in contrast to pathogen eliminating (Qu-Xie) herbs. However, the commonness in the biological basis of health-strengthening herbs remains to be holistically elucidated. In this study, an innovative high-throughput research strategy integrating computational and experimental methods of network pharmacology was proposed, and 22 health-strengthening herbs were selected for the investigation. Additionally, 25 pathogen-eliminating herbs were included for comparison. First, based on network-based, large-scale target prediction, we analyzed the target profiles of 1446 TCM compounds. Next, the actions of 166 compounds on 420 antitumor or immune-related genes were measured using a unique high-throughput screening strategy by high-throughput sequencing, referred to as HTS2. Furthermore, the structural information and the antitumor activity of the compounds in health-strengthening and pathogen-eliminating herbs were compared. Using network pharmacology analysis, we discovered that: (1) Functionally, the predicted targets of compounds from health strengthening herbs were enriched in both immune-related and antitumor pathways, similar to those of pathogen eliminating herbs. As a case study, galloylpaeoniflorin, a compound in a health strengthening herb Radix Paeoniae Alba (Bai Shao), was found to exert antitumor effects both in vivo and in vitro. Yet the inhibitory effects of the compounds from pathogen eliminating herbs on tumor cells proliferation as a whole were significantly stronger than those in health-strengthening herbs (p < 0.001). Moreover, the percentage of assay compounds in health-strengthening herbs with the predicted targets enriched in the immune-related pathways (e.g., natural killer cell mediated cytotoxicity and antigen processing and presentation) were significantly higher than that in pathogen-eliminating herbs (p < 0.05). This finding was supported by the immune-enhancing effects of a group of compounds from health-strengthening herbs indicated by differentially expressed genes in the HTS2 results. (2) Compounds in the same herb may exhibit the same or distinguished mechanisms in cancer treatment, which was demonstrated as the compounds influence pathway gene expressions in the same or opposite directions. For example, acetyl ursolic acid and specnuezhenide in a health-strengthening herb Fructus Ligustri lucidi (Nv Zhen Zi) both upregulated gene expressions in T cell receptor signaling pathway. Together, this study suggested greater potentials in tumor immune microenvironment regulation and tumor prevention than in direct killing tumor cells of health-strengthening herbs generally, and provided a systematic strategy for unveiling the commonness in the biological basis of health-strengthening herbs in cancer treatment. Full article
(This article belongs to the Special Issue Application of Bioinformatics in Cancers)
Figures

Graphical abstract

Open AccessArticle Predicting 90-Day Mortality in Locoregionally Advanced Head and Neck Squamous Cell Carcinoma after Curative Surgery
Cancers 2018, 10(10), 392; https://doi.org/10.3390/cancers10100392
Received: 4 September 2018 / Revised: 13 October 2018 / Accepted: 18 October 2018 / Published: 22 October 2018
PDF Full-text (652 KB) | HTML Full-text | XML Full-text | Supplementary Files
Abstract
Purpose: To propose a risk classification scheme for locoregionally advanced (Stages III and IV) head and neck squamous cell carcinoma (LA-HNSCC) by using the Wu comorbidity score (WCS) to quantify the risk of curative surgeries, including tumor resection and radical neck dissection. Methods: [...] Read more.
Purpose: To propose a risk classification scheme for locoregionally advanced (Stages III and IV) head and neck squamous cell carcinoma (LA-HNSCC) by using the Wu comorbidity score (WCS) to quantify the risk of curative surgeries, including tumor resection and radical neck dissection. Methods: This study included 55,080 patients with LA-HNSCC receiving curative surgery between 2006 and 2015 who were identified from the Taiwan Cancer Registry database; the patients were classified into two groups, mortality (n = 1287, mortality rate = 2.34%) and survival (n = 53,793, survival rate = 97.66%), according to the event of mortality within 90 days of surgery. Significant risk factors for mortality were identified using a stepwise multivariate Cox proportional hazards model. The WCS was calculated using the relative risk of each risk factor. The accuracy of the WCS was assessed using mortality rates in different risk strata. Results: Fifteen comorbidities significantly increased mortality risk after curative surgery. The patients were divided into low-risk (WCS, 0–6; 90-day mortality rate, 0–1.57%), intermediate-risk (7–11; 2.71–9.99%), high-risk (12–16; 17.30–20.00%), and very-high-risk (17–18 and >18; 46.15–50.00%) strata. The 90-day survival rates were 98.97, 95.85, 81.20, and 53.13% in the low-, intermediate-, high-, and very-high-risk patients, respectively (log-rank p < 0.0001). The five-year overall survival rates after surgery were 70.86, 48.62, 22.99, and 18.75% in the low-, intermediate-, high-, and very-high-risk patients, respectively (log-rank p < 0.0001). Conclusion: The WCS is an accurate tool for assessing curative-surgery-related 90-day mortality risk and overall survival in patients with LA-HNSCC. Full article
(This article belongs to the Special Issue Application of Bioinformatics in Cancers)
Figures

Figure 1

Review

Jump to: Research, Other

Open AccessReview A Review on a Deep Learning Perspective in Brain Cancer Classification
Cancers 2019, 11(1), 111; https://doi.org/10.3390/cancers11010111
Received: 29 November 2018 / Revised: 7 January 2019 / Accepted: 10 January 2019 / Published: 18 January 2019
PDF Full-text (5628 KB) | HTML Full-text | XML Full-text
Abstract
A World Health Organization (WHO) Feb 2018 report has recently shown that mortality rate due to brain or central nervous system (CNS) cancer is the highest in the Asian continent. It is of critical importance that cancer be detected earlier so that many [...] Read more.
A World Health Organization (WHO) Feb 2018 report has recently shown that mortality rate due to brain or central nervous system (CNS) cancer is the highest in the Asian continent. It is of critical importance that cancer be detected earlier so that many of these lives can be saved. Cancer grading is an important aspect for targeted therapy. As cancer diagnosis is highly invasive, time consuming and expensive, there is an immediate requirement to develop a non-invasive, cost-effective and efficient tools for brain cancer characterization and grade estimation. Brain scans using magnetic resonance imaging (MRI), computed tomography (CT), as well as other imaging modalities, are fast and safer methods for tumor detection. In this paper, we tried to summarize the pathophysiology of brain cancer, imaging modalities of brain cancer and automatic computer assisted methods for brain cancer characterization in a machine and deep learning paradigm. Another objective of this paper is to find the current issues in existing engineering methods and also project a future paradigm. Further, we have highlighted the relationship between brain cancer and other brain disorders like stroke, Alzheimer’s, Parkinson’s, and Wilson’s disease, leukoriaosis, and other neurological disorders in the context of machine learning and the deep learning paradigm. Full article
(This article belongs to the Special Issue Application of Bioinformatics in Cancers)
Figures

Figure 1

Other

Jump to: Research, Review

Open AccessBrief Report Breast Cancer Prognosis Using a Machine Learning Approach
Cancers 2019, 11(3), 328; https://doi.org/10.3390/cancers11030328
Received: 18 December 2018 / Revised: 26 February 2019 / Accepted: 4 March 2019 / Published: 7 March 2019
PDF Full-text (476 KB) | HTML Full-text | XML Full-text
Abstract
Machine learning (ML) has been recently introduced to develop prognostic classification models that can be used to predict outcomes in individual cancer patients. Here, we report the significance of an ML-based decision support system (DSS), combined with random optimization (RO), to extract prognostic [...] Read more.
Machine learning (ML) has been recently introduced to develop prognostic classification models that can be used to predict outcomes in individual cancer patients. Here, we report the significance of an ML-based decision support system (DSS), combined with random optimization (RO), to extract prognostic information from routinely collected demographic, clinical and biochemical data of breast cancer (BC) patients. A DSS model was developed in a training set (n = 318), whose performance analysis in the testing set (n = 136) resulted in a C-index for progression-free survival of 0.84, with an accuracy of 86%. Furthermore, the model was capable of stratifying the testing set into two groups of patients with low- or high-risk of progression with a hazard ratio (HR) of 10.9 (p < 0.0001). Validation in multicenter prospective studies and appropriate management of privacy issues in relation to digital electronic health records (EHR) data are presently needed. Nonetheless, we may conclude that the implementation of ML algorithms and RO models into EHR data might help to achieve prognostic information, and has the potential to revolutionize the practice of personalized medicine. Full article
(This article belongs to the Special Issue Application of Bioinformatics in Cancers)
Figures

Figure 1

Cancers EISSN 2072-6694 Published by MDPI AG, Basel, Switzerland RSS E-Mail Table of Contents Alert
Back to Top