ChiPPI: a novel method for mapping chimeric protein-protein interactions uncovers selection principles of protein fusion events in cancer

FusionPDB: a knowledgebase of human fusion proteins

Tumorigenic functions due to the formation of fusion genes have been targeted for cancer therapeutics (i.e. kinase inhibitors). However, many fusion proteins involved in various cellular processes have not been studied for targeted therapeutics. This is because the lack of complete fusion protein sequences and their whole 3D structures has made it challenging to develop new therapeutic strategies. To fill these critical gaps, we developed a computational pipeline and a resource of human fusion proteins named FusionPDB, available at https://compbio.uth.edu/FusionPDB. FusionPDB is organized into four levels: 43K fusion protein sequences (14.7K in-frame fusion genes, Level 1), over 2300 + 1267 fusion protein 3D structures (from 2300 recurrent and 266 manually curated in-frame fusion genes, Level 2), pLDDT score analysis for the 1267 fusion proteins from 266 manually curated fusion genes (Level 3), and virtual screening outcomes for 68 selected fusion proteins from 266 manually curated fusion genes (Level 4). FusionPDB is the only resource providing whole 3D structures of fusion proteins and comprehensive knowledge of human fusion proteins. It will be regularly updated until it covers all human fusion proteins in the future.

Functional and regulatory impact of chimeric RNAs in human normal and cancer cells

Fusions of two genes can lead to the generation of chimeric RNAs, which may have a distinct functional role from their original molecules. Chimeric RNAs could encode novel functional proteins or serve as novel long noncoding RNAs (lncRNAs). The appearance of chimeric RNAs in a cell could help to generate new functionality and phenotypic diversity that might facilitate this cell to survive against new environmental stress. Several recent studies have demonstrated the functional roles of various chimeric RNAs in cancer progression and are considered as biomarkers for cancer diagnosis and sometimes even drug targets. Further, the growing evidence demonstrated the potential functional association of chimeric RNAs with cancer heterogeneity and drug resistance cancer evolution. Recent studies highlighted that chimeric RNAs also have functional potentiality in normal physiological processes. Several functionally potential chimeric RNAs were discovered in human cancer and normal cells in the last two decades. This could indicate that chimeric RNAs are the hidden layer of the human transcriptome that should be explored from the functional insights to better understand the functional evolution of the genome and disease development that could facilitate clinical practice improvements. This review summarizes the current knowledge of chimeric RNAs and highlights their functional, regulatory, and evolutionary impact on different cancers and normal physiological processes. Further, we will discuss the potential functional roles of a recently discovered novel class of chimeric RNAs named sense-antisense/cross-strand chimeric RNAs generated by the fusion of the bi-directional transcripts of the same gene. This article is categorized under: Regulatory RNAs/RNAi/Riboswitches > Regulatory RNAs.

Interactomics in plant defence: progress and opportunities

Interactomics is a branch of systems biology that deals with the study of protein-protein interactions and how these interactions influence phenotypes. Identifying the interactomes involved during host-pathogen interaction events may bring us a step closer to deciphering the molecular mechanisms underlying plant defence. Here, we conducted a systematic review of plant interactomics studies over the last two decades and found that while a substantial progress has been made in the field, plant-pathogen interactomics remains a less-travelled route. As an effort to facilitate the progress in this field, we provide here a comprehensive research pipeline for an in planta plant-pathogen interactomics study that encompasses the in silico prediction step to the validation step, unconfined to model plants. We also highlight four challenges in plant-pathogen interactomics with plausible solution(s) for each.

The Landscape of Expressed Chimeric Transcripts in the Blood of Severe COVID-19 Infected Patients

The ongoing COVID-19 pandemic caused by SARS-CoV-2 infections has quickly developed into a global public health threat. COVID-19 patients show distinct clinical features, and in some cases, during the severe stage of the condition, the disease severity leads to an acute respiratory disorder. In spite of several pieces of research in this area, the molecular mechanisms behind the development of disease severity are still not clearly understood. Recent studies demonstrated that SARS-CoV-2 alters the host cell splicing and transcriptional response to overcome the host immune response that provides the virus with favorable conditions to replicate efficiently within the host cells. In several disease conditions, aberrant splicing could lead to the development of novel chimeric transcripts that could promote the functional alternations of the cell. As severe SARS-CoV-2 infection was reported to cause abnormal splicing in the infected cells, we could expect the generation and expression of novel chimeric transcripts. However, no study so far has attempted to check whether novel chimeric transcripts are expressed in severe SARS-CoV-2 infections. In this study, we analyzed several publicly available blood transcriptome datasets of severe COVID-19, mild COVID-19, other severe respiratory viral infected patients, and healthy individuals. We identified 424 severe COVID-19 -specific chimeric transcripts, 42 of which were recurrent. Further, we detected 189 chimeric transcripts common to severe COVID-19 and multiple severe respiratory viral infections. Pathway and gene enrichment analysis of the parental genes of these two subsets of chimeric transcripts reveals that these are potentially involved in immune-related processes, interferon signaling, and inflammatory responses, which signify their potential association with immune dysfunction leading to the development of disease severity. Our study provides the first detailed expression landscape of chimeric transcripts in severe COVID-19 and other severe respiratory viral infections.

A review of bioinformatics tools and web servers in different microarray platforms used in cancer research

Over the past decade, conventional lab work strategies have gradually shifted from being limited to a laboratory setting towards a bioinformatics era to help manage and process the vast amounts of data generated by omics technologies. The present work outlines the latest contributions of bioinformatics in analyzing microarray data and their application to cancer. We dissect different microarray platforms and their use in gene expression in cancer models. We highlight how computational advances empowered the microarray technology in gene expression analysis. The study on protein-protein interaction databases classified into primary, derived, meta-database, and prediction databases describes the strategies to curate and predict novel interaction networks in silico. In addition, we summarize the areas of bioinformatics where neural graph networks are currently being used, such as protein functions, protein interaction prediction, and in silico drug discovery and development. We also discuss the role of deep learning as a potential tool in the prognosis, diagnosis, and treatment of cancer. Integrating these resources efficiently, practically, and ethically is likely to be the most challenging task for the healthcare industry over the next decade; however, we believe that it is achievable in the long term.

Fusion proteins mediate alternation of protein interaction networks in cancers

Fusions of two different genes could lead to the production of chimeric RNAs, which could be translated into novel fusion (or chimeric) proteins. Fusion proteins often act as oncoproteins and drive cancer development, particularly in leukemia and lymphomas. Fusion proteins modify the existing protein-protein interaction (PPI) networks, which could eliminate some PPIs by removing protein domains in such fusions. This alternation of protein interaction networks could impact the signaling pathways and switch on the cancer-promoting activity that could drive the generation of cancer phenotypes and/or loss of controlled apoptosis. Thus, knowledge of the fusion proteins and their protein interaction networks could facilitate a deeper molecular understanding of cancer development, which could help to design new approaches for cancer therapies. Here, we discuss the structural features of fusion proteins and how they impact the PPI networks in cancers. Further, we discuss how to analyze the fusion protein-mediated alternation of PPI networks in cancers.

The Landscape of Novel Expressed Chimeric RNAs in Rheumatoid Arthritis

In cancers and other complex diseases, the fusion of two genes can lead to the production of chimeric RNAs, which are associated with disease development. Several recurrent chimeric RNAs are expressed in different cancers and are thus used for clinical cancer diagnosis. Rheumatoid arthritis (RA) is an immune-mediated joint disorder resulting in synovial inflammation and joint destruction. Despite advances in therapy, many patients do not respond to treatment and present persistent inflammation. Understanding the landscape of chimeric RNA expression in RA patients could provide a better insight into RA pathogenesis, which might provide better treatment strategies and tailored therapies. Accordingly, we analyzed the publicly available RNA-seq data of synovium tissue from 151 RA patients and 28 healthy controls and were able to identify 37 recurrent chimeric RNAs found to be expressed in at least 3 RA samples. Furthermore, the parental genes of these 37 recurrent chimeric RNAs were found to be differentially expressed and enriched in immune-related processes, such as adaptive immune response and the positive regulation of B-cell activation. Interestingly, the appearance of 5 coding and 23 non-coding chimeric RNAs might be associated with regulating their parental gene expression, leading to the generation of dysfunctional immune responses, such as inflammation and bone destruction. Therefore, in this paper, we present the first study to demonstrate the novel chimeric RNAs that are highly expressed and functional in RA.

Detection of gene mutations and gene-gene fusions in circulating cell-free DNA of glioblastoma patients: an avenue for clinically relevant diagnostic analysis

Glioblastoma (GBM) is the most common type of glioma and is uniformly fatal. Currently, tumour heterogeneity and mutation acquisition are major impedances for tailoring personalized therapy. We collected blood and tumour tissue samples from 25 GBM patients and 25 blood samples from healthy controls. Cell‐free DNA (cfDNA) was extracted from the plasma of GBM patients and from healthy controls. Tumour DNA was extracted from fresh tumour samples. Extracted DNA was sequenced using a whole‐genome sequencing procedure. We also collected 180 tumour DNA datasets from GBM patients publicly available at the TCGA/PANCANCER project. These data were analysed for mutations and gene–gene fusions that could be potential druggable targets. We found that plasma cfDNA concentrations in GBM patients were significantly elevated (22.6 ± 5 ng·mL⁻¹), as compared to healthy controls (1.4 ± 0.4 ng·mL⁻¹) of the same average age. We identified unique mutations in the cfDNA and tumour DNA of each GBM patient, including some of the most frequently mutated genes in GBM according to the COSMIC database (TP53, 18.75%; EGFR, 37.5%; NF1, 12.5%; LRP1B, 25%; IRS4, 25%). Using our gene–gene fusion database, ChiTaRS 5.0, we identified gene–gene fusions in cfDNA and tumour DNA, such as KDR–PDGFRA and NCDN–PDGFRA, which correspond to previously reported alterations of PDGFRA in GBM (44% of all samples). Interestingly, the PDGFRA protein fusions can be targeted by tyrosine kinase inhibitors such as imatinib, sunitinib, and sorafenib. Moreover, we identified BCR–ABL1 (in 8% of patients), COL1A1–PDGFB (8%), NIN–PDGFRB (8%), and FGFR1–BCR (4%) in cfDNA of patients, which can be targeted by analogues of imatinib. ROS1 fusions (CEP85L–ROS1 and GOPC–ROS1), identified in 8% of patient cfDNA, might be targeted by crizotinib, entrectinib, or larotrectinib. Thus, our study suggests that integrated analysis of cfDNA plasma concentration, gene mutations, and gene–gene fusions can serve as a diagnostic modality for distinguishing GBM patients who may benefit from targeted therapy. These results open new avenues for precision medicine in GBM, using noninvasive liquid biopsy diagnostics to assess personalized patient profiles. Moreover, repeated detection of druggable targets over the course of the disease may provide real‐time information on the evolving molecular landscape of the tumour.

Emerging Role of Chimeric RNAs in Cell Plasticity and Adaptive Evolution of Cancer Cells

Gene fusions can give rise to somatic alterations in cancers. Fusion genes have the potential to create chimeric RNAs, which can generate the phenotypic diversity of cancer cells, and could be associated with novel molecular functions related to cancer cell survival and proliferation. The expression of chimeric RNAs in cancer cells might impact diverse cancer-related functions, including loss of apoptosis and cancer cell plasticity, and promote oncogenesis. Due to their recurrence in cancers and functional association with oncogenic processes, chimeric RNAs are considered biomarkers for cancer diagnosis. Several recent studies demonstrated that chimeric RNAs could lead to the generation of new functionality for the resistance of cancer cells against drug therapy. Therefore, targeting chimeric RNAs in drug resistance cancer could be useful for developing precision medicine. So, understanding the functional impact of chimeric RNAs in cancer cells from an evolutionary perspective will be helpful to elucidate cancer evolution, which could provide a new insight to design more effective therapies for cancer patients in a personalized manner.

EvoProDom: Evolutionary modeling of protein families by assessing translocations of protein domains

Here, we introduce a novel ‘evolution of protein domains’ (EvoProDom) model for describing the evolution of proteins based on the ‘mix and merge’ of protein domains. We assembled and integrated genomic and proteomic data comprising protein domain content and orthologous proteins from 109 organisms. In EvoProDom, we characterized evolutionary events, particularly, translocations, as reciprocal exchanges of protein domains between orthologous proteins in different organisms. We showed that protein domains that translocate with highly frequency are generated by transcripts enriched in trans‐splicing events, that is, the generation of novel transcripts from the fusion of two distinct genes. In EvoProDom, we describe a general method to collate orthologous protein annotation from KEGG, and protein domain content from protein sequences using tools such as KoFamKOAL and Pfam. To summarize, EvoProDom presents a novel model for protein evolution based on the ‘mix and merge’ of protein domains rather than DNA‐based evolution models. This confers the advantage of considering chromosomal alterations as drivers of protein evolutionary events.

Design principles of gene evolution for niche adaptation through changes in protein-protein interaction networks

In contrast to fossorial and above-ground organisms, subterranean species have adapted to the extreme stresses of living underground. We analyzed the predicted protein–protein interactions (PPIs) of all gene products, including those of stress-response genes, among nine subterranean, ten fossorial, and 13 aboveground species. We considered 10,314 unique orthologous protein families and constructed 5,879,879 PPIs in all organisms using ChiPPI. We found strong association between PPI network modulation and adaptation to specific habitats, noting that mutations in genes and changes in protein sequences were not linked directly with niche adaptation in the organisms sampled. Thus, orthologous hypoxia, heat-shock, and circadian clock proteins were found to cluster according to habitat, based on PPIs rather than on sequence similarities. Curiously, "ordered" domains were preserved in aboveground species, while "disordered" domains were conserved in subterranean organisms, and confirmed for proteins in DistProt database. Furthermore, proteins with disordered regions were found to adopt significantly less optimal codon usage in subterranean species than in fossorial and above-ground species. These findings reveal design principles of protein networks by means of alterations in protein domains, thus providing insight into deep mechanisms of evolutionary adaptation, generally, and particularly of species to underground living and other confined habitats.

The landscape of chimeric RNAs in non-diseased tissues and cells

Chimeric RNAs and their encoded proteins have been traditionally viewed as unique features of neoplasia, and have been used as biomarkers and therapeutic targets for multiple cancers. Recent studies have demonstrated that chimeric RNAs also exist in non-cancerous cells and tissues, although large-scale, genome-wide studies of chimeric RNAs in non-diseased tissues have been scarce. Here, we explored the landscape of chimeric RNAs in 9495 non-diseased human tissue samples of 53 different tissues from the GTEx project. Further, we established means for classifying chimeric RNAs, and observed enrichment for particular classifications as more stringent filters are applied. We experimentally validated a subset of chimeric RNAs from each classification and demonstrated functional relevance of two chimeric RNAs in non-cancerous cells. Importantly, our list of chimeric RNAs in non-diseased tissues overlaps with some entries in several cancer fusion databases, raising concerns for some annotations. The data from this study provides a large repository of chimeric RNAs present in non-diseased tissues, which can be used as a control dataset to facilitate the identification of true cancer-specific chimeras.

Targeted Identification of Protein Interactions in Eukaryotic mRNA Translation

To identify protein-protein interactions and phosphorylated amino acid sites in eukaryotic mRNA translation, replicate TAP-MudPIT and control experiments are performed targeting Saccharomyces cerevisiae genes previously implicated in eukaryotic mRNA translation by their genetic and/or functional roles in translation initiation, elongation, termination, or interactions with ribosomal complexes. Replicate tandem affinity purifications of each targeted yeast TAP-tagged mRNA translation protein coupled with multidimensional liquid chromatography and tandem mass spectrometry analysis are used to identify and quantify copurifying proteins. To improve sensitivity and minimize spurious, nonspecific interactions, a novel cross-validation approach is employed to identify the most statistically significant protein-protein interactions. Using experimental and computational strategies discussed herein, the previously described protein composition of the canonical eukaryotic mRNA translation initiation, elongation, and termination complexes is calculated. In addition, statistically significant unpublished protein interactions and phosphorylation sites for S. cerevisiae's mRNA translation proteins and complexes are identified.

Pathway perturbations in signaling networks: Linking genotype to phenotype

Genes and gene products interact with each other to form signal transduction networks in the cell. The interactome networks are under intricate regulation in physiological conditions, but could go awry upon genome instability caused by genetic mutations. In the past decade with next-generation sequencing technologies, an increasing number of genomic mutations have been identified in a variety of disease patients and healthy individuals. As functional and systematic studies on these mutations leap forward, they begin to reveal insights into cellular homeostasis and disease mechanisms. In this review, we discuss recent advances in the field of network biology and signaling pathway perturbations upon genomic changes, and highlight the success of various omics datasets in unraveling genotype-to-phenotype relationships.

ChiTaRS 5.0: the comprehensive database of chimeric transcripts matched with druggable fusions and 3D chromatin maps

Chimeric RNA transcripts are formed when exons from two genes fuse together, often due to chromosomal translocations, transcriptional errors or trans-splicing effect. While these chimeric RNAs produce functional proteins only in certain cases, they play a significant role in disease phenotyping and progression. ChiTaRS 5.0 (http://chitars.md.biu.ac.il/) is the latest and most comprehensive chimeric transcript repository, with 111 582 annotated entries from eight species, including 23 167 known human cancer breakpoints. The database includes unique information correlating chimeric breakpoints with 3D chromatin contact maps, generated from public datasets of chromosome conformation capture techniques (Hi-C). In this update, we have added curated information on druggable fusion targets matched with chimeric breakpoints, which are applicable to precision medicine in cancers. The introduction of a new section that lists chimeric RNAs in various cell-lines is another salient feature. Finally, using text-mining techniques, novel chimeras in Alzheimer's disease, schizophrenia, dyslexia and other diseases were collected in ChiTaRS. Thus, this improved version is an extensive catalogue of chimeras from multiple species. It extends our understanding of the evolution of chimeric transcripts in eukaryotes and contributes to the analysis of 3D genome conformational changes and the functional role of chimeras in the etiopathogenesis of cancers and other complex diseases.

Identification of Chimeric RNAs Using RNA-Seq Reads and Protein-Protein Interactions of Translated Chimeras

Chimeric RNA moieties typically consist of exons from two genes expressed from different genomic locations and produced by chromosomal translocations, trans-splicing or transcription errors. Recent advances in next-generation sequencing procedures have opened new horizons for identification of novel chimeric transcripts in various diseases in a personalized manner. Here we describe the detailed computational procedures to identify chimeric transcripts using RNA-seq reads. Moreover, we elaborate on the domain-domain co-occurrence method to detect alterations in chimeric protein-protein interaction (ChiPPI) networks produced by chimeric RNA that are translated to chimeric proteins.

Genetics' Piece of the PI: Inferring the Origin of Complex Traits and Diseases from Proteome-Wide Protein-Protein Interaction Dynamics

How do common and rare genetic polymorphisms contribute to quantitative traits or disease risk and progression? Multiple human traits have been extensively characterized at the genomic level, revealing their complex genetic architecture. However, it is difficult to resolve the mechanisms by which specific variants contribute to a phenotype. Recently, analyses of variant effects on molecular traits have uncovered intermediate mechanisms that link sequence variation to phenotypic changes. Yet, these methods only capture a fraction of genetic contributions to phenotype. Here, in reviewing the field, it is proposed that complex traits can be understood by characterizing the dynamics of biochemical networks within living cells, and that the effects of genetic variation can be captured on these networks by using protein-protein interaction (PPI) methodologies. This synergy between PPI methodologies and the genetics of complex traits opens new avenues to investigate the molecular etiology of human diseases and to facilitate their prevention or treatment.

ProtFus: A Comprehensive Method Characterizing Protein-Protein Interactions of Fusion Proteins

Tailored therapy aims to cure cancer patients effectively and safely, based on the complex interactions between patients' genomic features, disease pathology and drug metabolism. Thus, the continual increase in scientific literature drives the need for efficient methods of data mining to improve the extraction of useful information from texts based on patients' genomic features. An important application of text mining to tailored therapy in cancer encompasses the use of mutations and cancer fusion genes as moieties that change patients' cellular networks to develop cancer, and also affect drug metabolism. Fusion proteins, which are derived from the slippage of two parental genes, are produced in cancer by chromosomal aberrations and trans-splicing. Given that the two parental proteins for predicted fusion proteins are known, we used our previously developed method for identifying chimeric protein-protein interactions (ChiPPIs) associated with the fusion proteins. Here, we present a validation approach that receives fusion proteins of interest, predicts their cellular network alterations by ChiPPI and validates them by our new method, ProtFus, using an online literature search. This process resulted in a set of 358 fusion proteins and their corresponding protein interactions, as a training set for a Naïve Bayes classifier, to identify predicted fusion proteins that have reliable evidence in the literature and that were confirmed experimentally. Next, for a test group of 1817 fusion proteins, we were able to identify from the literature 2908 PPIs in total, across 18 cancer types. The described method, ProtFus, can be used for screening the literature to identify unique cases of fusion proteins and their PPIs, as means of studying alterations of protein networks in cancers. Availability: http://protfus.md.biu.ac.il/.

The chromatin remodeler Chd1 regulates cohesin in budding yeast and humans

Chd1 is a chromatin remodeler that is involved in nucleosome positioning and transcription. Deletion of CHD1 is a frequent event in prostate cancer. The Structural Maintenance of Chromosome (SMC) complex cohesin mediates long-range chromatin interactions and is involved in maintaining genome stability. We provide new evidence that Chd1 is a regulator of cohesin. In the yeast S. cerevisiae, Chd1 is not essential for viability. We show that deletion of the gene leads to a defect in sister chromatid cohesion and in chromosome morphology. Chl1 is a non-essential DNA helicase that has been shown to regulate cohesin loading. Surprisingly, co-deletion of CHD1 and CHL1 results in an additive cohesion defect but partial suppression of the chromosome structure phenotype. We found that the cohesin regulator Pds5 is overexpressed when Chd1 and Chl1 are deleted. However, Pds5 expression is reduced to wild type levels when both genes are deleted. Finally, we show a correlation in the expression of CHD1 and cohesin genes in prostate cancer patients. Furthermore, we show that overexpression of cohesin subunits is correlated with the aggressiveness of the tumor. The biological roles of the interplay between Chd1, Chl1 and SMCs are discussed.

Altered expression of metabolic pathways in CLL detected by unlabelled quantitative mass spectrometry analysis

Chronic lymphocytic leukaemia (CLL) remains the most common incurable malignancy of B cells in the western world. Patient outcomes are heterogeneous and can be difficult to predict with current prognostic markers. Here, we used a quantitative label-free proteomic technique to ascertain differences in the B-cell proteome from healthy donors and CLL patients with either mutated (M-CLL) or unmutated (UM-CLL) IGHV to identify new prognostic markers. In peripheral B-CLL cells, 349 (22%) proteins were differentially expressed between normal B cells and B-CLL cells and 189 (12%) were differentially expressed between M-CLL and UM-CLL. We also examined the proteome of proliferating CLL cells in the lymph nodes, and identified 76 (~8%) differentially expressed proteins between healthy and CLL lymph nodes. B-CLL cells show over-expression of proteins involved in lipid and cholesterol metabolism. A comprehensive lipidomic analysis highlighted large differences in glycolipids and sphingolipids. A shift was observed from the pro-apoptotic lipid ceramide towards the anti-apoptotic/chemoresistant lipid, glucosylceramide, which was more evident in patients with aggressive disease (UM-CLL). This study details a novel quantitative proteomic technique applied for the first time to primary patient samples in CLL and highlights that primary CLL lymphocytes display markers of a metabolic shift towards lipid synthesis and breakdown.

FusionGDB: fusion gene annotation DataBase

Gene fusion is one of the hallmarks of cancer genome via chromosomal rearrangement initiated by DNA double-strand breakage. To date, many fusion genes (FGs) have been established as important biomarkers and therapeutic targets in multiple cancer types. To better understand the function of FGs in cancer types and to promote the discovery of clinically relevant FGs, we built FusionGDB (Fusion Gene annotation DataBase) available at https://ccsm.uth.edu/FusionGDB. We collected 48 117 FGs across pan-cancer from three representative fusion gene resources: the improved database of chimeric transcripts and RNA-seq data (ChiTaRS 3.1), an integrative resource for cancer-associated transcript fusions (TumorFusions), and The Cancer Genome Atlas (TCGA) fusions by Gao et al. For these ∼48K FGs, we performed functional annotations including gene assessment across pan-cancer fusion genes, open reading frame (ORF) assignment, and retention search of 39 protein features based on gene structures of multiple isoforms with different breakpoints. We also provided the fusion transcript and amino acid sequences according to multiple breakpoints and transcript isoforms. Our analyses identified 331, 303 and 667 in-frame FGs with retaining kinase, DNA-binding, and epigenetic factor domains, respectively, as well as 976 FGs lost protein-protein interaction. FusionGDB provides six categories of annotations: FusionGeneSummary, FusionProtFeature, FusionGeneSequence, FusionGenePPI, RelatedDrug and RelatedDisease.

Synstable Fusion: A Network-Based Algorithm for Estimating Driver Genes in Fusion Structures

Gene fusion structure is a class of common somatic mutational events in cancer genomes, which are often formed by chromosomal mutations. Identifying the driver gene(s) in a fusion structure is important for many downstream analyses and it contributes to clinical practices. Existing computational approaches have prioritized the importance of oncogenes by incorporating prior knowledge from gene networks. However, different methods sometimes suffer different weaknesses when handling gene fusion data due to multiple issues such as fusion gene representation, network integration, and the effectiveness of the evaluation algorithms. In this paper, Synstable Fusion (SYN), an algorithm for computationally evaluating the fusion genes, is proposed. This algorithm uses network-based strategy by incorporating gene networks as prior information, but estimates the driver genes according to the destructiveness hypothesis. This hypothesis balances the two popular evaluation strategies in the existing studies, thereby providing more comprehensive results. A machine learning framework is introduced to integrate multiple networks and further solve the conflicting results from different networks. In addition, a synchronous stability model is established to reduce the computational complexity of the evaluation algorithm. To evaluate the proposed algorithm, we conduct a series of experiments on both artificial and real datasets. The results demonstrate that the proposed algorithm performs well on different configurations and is robust when altering the internal parameter settings.

A Novel Approach for Identifying Relevant Genes for Breast Cancer Survivability on Specific Therapies

Analyzing the genetic activity of breast cancer survival for a specific type of therapy provides a better understanding of the body response to the treatment and helps select the best course of action and while leading to the design of drugs based on gene activity. In this work, we use supervised and nonsupervised machine learning methods to deal with a multiclass classification problem in which we label the samples based on the combination of the 5-year survivability and treatment; we focus on hormone therapy, radiotherapy, and surgery. The proposed nonsupervised hierarchical models are created to find the highest separability between combinations of the classes. The supervised model consists of a combination of feature selection techniques and efficient classifiers used to find a potential set of biomarker genes specific to response to therapy. The results show that different models achieve different performance scores with accuracies ranging from 80.9% to 100%. We have investigated the roles of many biomarkers through the literature and found that some of the discriminative genes in the computational model such as ZC3H11A, VAX2, MAF1, and ZFP91 are related to breast cancer and other types of cancer.

FusionPathway: Prediction of pathways and therapeutic targets associated with gene fusions in cancer

Numerous gene fusions have been uncovered across multiple cancer types. Although the ability to target several of these fusions has led to the development of some successful anti-cancer drugs, most of them are not druggable. Understanding the molecular pathways of a fusion is important in determining its function in oncogenesis and in developing therapeutic strategies for patients harboring the fusion. However, the molecular pathways have been elucidated for only a few fusions, in part because of the labor-intensive nature of the required functional assays. Therefore, we developed a domain-based network approach to infer the pathways of a fusion. Molecular interactions of a fusion are first predicted by using its protein domain composition, and its associated pathways are then inferred from these molecular interactions. We demonstrated the capabilities of this approach by primarily applying it to the well-studied BCR-ABL1 fusion. The approach was also applied to two undruggable fusions in sarcoma, EWS-FL1 and FUS-DDIT3. We successfully identified known genes and pathways associated with these fusions and satisfactorily validated these predictions using several benchmark sets. The predictions of EWS-FL1 and FUS-DDIT3 also correlate with results of high-throughput drug screening. To our best knowledge, this is the first approach for inferring pathways of fusions.

We present a computational framework, FusionPathway, to infer the oncogenesis pathways of a fusion and help develop therapeutic strategies in these pathways for patients harboring the fusion. In this work, we successfully validated the capabilities of this approach through its application to the well-studied BCR-ABL1 fusion and two undruggable fusions in sarcoma, EWS-FL1 and FUS-DDIT3. Especially, the predictions of EWS-FL1 and FUS-DDIT3 correlate well with results of high-throughput drug screening in sarcoma cells. Therefore, FusionPathway can be an effective method to infer pathways and potential therapeutic targets that are associated with those undruggable fusions. Our results of BCR-ABL1 also suggest that FusionPathway may be able to help elucidate pathway-dependent mechanisms of resistances to those kinase fusion-targeting therapies and develop strategies to overcome the resistances. In addition, the developed R package of FusionPathways (https://github.com/perwu/FusionPathway/) can help people easily apply our approach to study other important fusions in cancer.

Analyses of Protein Interaction Networks Using Computational Tools

The knowledge of protein-protein interactions (PPIs) and PPI networks (PPINs) is the key to starting to understand the biological processes inside the cell. Many computational tools have been designed to help explore PPIs and PPINs, such as those for interaction detection, reliability assessment and interaction network construction. Here, the application of computational tools is reviewed from three perspectives: PPI database construction, PPI prediction, and interaction network construction and analysis. This overview will provide researchers guidance on choosing appropriate methods for exploring PPIs.