Machine Learning Classification and Structure–Functional Analysis of Cancer Mutations Reveal Unique Dynamic and Network Signatures of Driver Sites in Oncogenes and Tumor Suppressor Genes

Systemic structural analysis of alterations reveals a common structural basis of driver mutations in cancer

A major effort in cancer research is to organize the complexities of the disease into fundamental traits. Despite conceptual progress in the last decades and the synthesis of hallmark features, no organizing principles governing cancer beyond cellular features exist. We analyzed experimentally determined structures harboring the most significant and prevalent driver missense mutations in human cancer, covering 73% (n = 168178) of the Catalog of Somatic Mutation in Cancer tumor samples (COSMIC). The results reveal that a single structural element-κ-helix (polyproline II helix)-lies at the core of driver point mutations, with significant enrichment in all major anatomical sites, suggesting that a small number of molecular traits are shared by most and perhaps all types of cancer. Thus, we uncovered the lowest possible level of organization at which carcinogenesis takes place at the protein level. This framework provides an initial scheme for a mechanistic understanding underlying the development of tumors and pinpoints key vulnerabilities.

Allostery: Allosteric Cancer Drivers and Innovative Allosteric Drugs

Here, we discuss the principles of allosteric activating mutations, propagation downstream of the signals that they prompt, and allosteric drugs, with examples from the Ras signaling network. We focus on Abl kinase where mutations shift the landscape toward the active, imatinib binding-incompetent conformation, likely resulting in the high affinity ATP outcompeting drug binding. Recent pharmacological innovation extends to allosteric inhibitor (GNF-5)-linked PROTAC, targeting Bcr-Abl1 myristoylation site, and broadly, allosteric heterobifunctional degraders that destroy targets, rather than inhibiting them. Designed chemical linkers in bifunctional degraders can connect the allosteric ligand that binds the target protein and the E3 ubiquitin ligase warhead anchor. The physical properties and favored conformational state of the engineered linker can precisely coordinate the distance and orientation between the target and the recruited E3. Allosteric PROTACs, noncompetitive molecular glues, and bitopic ligands, with covalent links of allosteric ligands and orthosteric warheads, increase the effective local concentration of productively oriented and placed ligands. Through covalent chemical or peptide linkers, allosteric drugs can collaborate with competitive drugs, degrader anchors, or other molecules of choice, driving innovative drug discovery.

Open Structural Data in Precision Medicine

Three-dimensional protein structural data at the molecular level are pivotal for successful precision medicine. Such data are crucial not only for discovering drugs that act to block the active site of the target mutant protein but also for clarifying to the patient and the clinician how the mutations harbored by the patient work. The relative paucity of structural data reflects their cost, challenges in their interpretation, and lack of clinical guidelines for their utilization. Rapid technological advancements in experimental high-resolution structural determination increasingly generate structures. Computationally, modeling algorithms, including molecular dynamics simulations, are becoming more powerful, as are compute-intensive hardware, particularly graphics processing units, overlapping with the inception of the exascale era. Accessible, freely available, and detailed structural and dynamical data can be merged with big data to powerfully transform personalized pharmacology. Here we review protein and emerging genome high-resolution data, along with means, applications, and examples underscoring their usefulness in precision medicine. Expected final online publication date for the Annual Review of Biomedical Data Science, Volume 5 is August 2022. Please see http://www.annualreviews.org/page/journal/pubdates for revised estimates.

Dissecting mutational allosteric effects in alkaline phosphatases associated with different Hypophosphatasia phenotypes: An integrative computational investigation

Hypophosphatasia (HPP) is a rare inherited disorder characterized by defective bone mineralization and is highly variable in its clinical phenotype. The disease occurs due to various loss-of-function mutations in ALPL, the gene encoding tissue-nonspecific alkaline phosphatase (TNSALP). In this work, a data-driven and biophysics-based approach is proposed for the large-scale analysis of ALPL mutations-from nonpathogenic to severe HPPs. By using a pipeline of synergistic approaches including sequence-structure analysis, network modeling, elastic network models and atomistic simulations, we characterized allosteric signatures and effects of the ALPL mutations on protein dynamics and function. Statistical analysis of molecular features computed for the ALPL mutations showed a significant difference between the control, mild and severe HPP phenotypes. Molecular dynamics simulations coupled with protein structure network analysis were employed to analyze the effect of single-residue variation on conformational dynamics of TNSALP dimers, and the developed machine learning model suggested that the topological network parameters could serve as a robust indicator of severe mutations. The results indicated that the severity of disease-associated mutations is often linked with mutation-induced modulation of allosteric communications in the protein. This study suggested that ALPL mutations associated with mild and more severe HPPs can exert markedly distinct effects on the protein stability and long-range network communications. By linking the disease phenotypes with dynamic and allosteric molecular signatures, the proposed integrative computational approach enabled to characterize and quantify the allosteric effects of ALPL mutations and role of allostery in the pathogenesis of HPPs.

Machine learning methods for prediction of cancer driver genes: a survey paper

Identifying the genes and mutations that drive the emergence of tumors is a critical step to improving our understanding of cancer and identifying new directions for disease diagnosis and treatment. Despite the large volume of genomics data, the precise detection of driver mutations and their carrying genes, known as cancer driver genes, from the millions of possible somatic mutations remains a challenge. Computational methods play an increasingly important role in discovering genomic patterns associated with cancer drivers and developing predictive models to identify these elements. Machine learning (ML), including deep learning, has been the engine behind many of these efforts and provides excellent opportunities for tackling remaining gaps in the field. Thus, this survey aims to perform a comprehensive analysis of ML-based computational approaches to identify cancer driver mutations and genes, providing an integrated, panoramic view of the broad data and algorithmic landscape within this scientific problem. We discuss how the interactions among data types and ML algorithms have been explored in previous solutions and outline current analytical limitations that deserve further attention from the scientific community. We hope that by helping readers become more familiar with significant developments in the field brought by ML, we may inspire new researchers to address open problems and advance our knowledge towards cancer driver discovery.

Protein domain-based approaches for the identification and prioritization of therapeutically actionable cancer variants

The tremendous number of cancer variants that can be detected by NGS analyses has required the development of computational approaches to prioritize mutations on the basis of their biological and clinical significance. Standard strategies take a gene-centric approach to the problem, allowing exclusively the identification of highly frequent variants. On the contrary, protein domain (PD)-based approaches allow to identify functionally relevant low frequency variants by searching for mutations that recur on analogous residues across homologous proteins (i.e. containing the same PD). Such approaches enable to transfer information about the effects and druggability from one known mutation to unknown ones. Here we describe how PD-based strategies work, and discuss how they could be exploited for mutation prioritization. The principle that mutations clustered on specific residues of PDs have the same functional consequences and are therapeutically actionable in a similar manner could help the choice of patient-specific targeted drugs, eventually improving the management of cancer patients.

Review of Machine Learning Technologies and Neural Networks in Drug Synergy Combination pharmacological research

Introduction: Nowadays an increase in the amount of information creates the need to replace and update data processing technologies. One of the tasks of clinical pharmacology is to create the right combination of drugs for the treatment of a particular disease. It takes months and even years to create a treatment regimen. Using machine learning (in silico) allows predicting how to get the right combination of drugs and skip the experimental steps in a study that take a lot of time and financial expenses. Gradual preparation is needed for the Deep Learning of Drug Synergy, starting from creating a base of drugs, their characteristics and ways of interacting.

Aim: Our review aims to draw attention to the prospect of the introduction of Deep Learning technology to predict possible combinations of drugs for the treatment of various diseases.

Materials and methods: Literary review of articles based on the PUBMED project and related bibliographic resources over the past 5 years (2015–2019).

Results and discussion: In the analyzed articles, Machine or Deep Learning completed the assigned tasks. It was able to determine the most appropriate combinations for the treatment of certain diseases, select the necessary regimen and doses. In addition, using this technology, new combinations have been identified that may be further involved in preclinical studies.

Conclusions: From the analysis of the articles, we obtained evidence of the positive effects of Deep Learning to select “key” combinations for further stages of preclinical research.

Allosteric Regulation at the Crossroads of New Technologies: Multiscale Modeling, Networks, and Machine Learning

Allosteric regulation is a common mechanism employed by complex biomolecular systems for regulation of activity and adaptability in the cellular environment, serving as an effective molecular tool for cellular communication. As an intrinsic but elusive property, allostery is a ubiquitous phenomenon where binding or disturbing of a distal site in a protein can functionally control its activity and is considered as the "second secret of life." The fundamental biological importance and complexity of these processes require a multi-faceted platform of synergistically integrated approaches for prediction and characterization of allosteric functional states, atomistic reconstruction of allosteric regulatory mechanisms and discovery of allosteric modulators. The unifying theme and overarching goal of allosteric regulation studies in recent years have been integration between emerging experiment and computational approaches and technologies to advance quantitative characterization of allosteric mechanisms in proteins. Despite significant advances, the quantitative characterization and reliable prediction of functional allosteric states, interactions, and mechanisms continue to present highly challenging problems in the field. In this review, we discuss simulation-based multiscale approaches, experiment-informed Markovian models, and network modeling of allostery and information-theoretical approaches that can describe the thermodynamics and hierarchy allosteric states and the molecular basis of allosteric mechanisms. The wealth of structural and functional information along with diversity and complexity of allosteric mechanisms in therapeutically important protein families have provided a well-suited platform for development of data-driven research strategies. Data-centric integration of chemistry, biology and computer science using artificial intelligence technologies has gained a significant momentum and at the forefront of many cross-disciplinary efforts. We discuss new developments in the machine learning field and the emergence of deep learning and deep reinforcement learning applications in modeling of molecular mechanisms and allosteric proteins. The experiment-guided integrated approaches empowered by recent advances in multiscale modeling, network science, and machine learning can lead to more reliable prediction of allosteric regulatory mechanisms and discovery of allosteric modulators for therapeutically important protein targets.

Why Are Some Driver Mutations Rare?

Understanding why driver mutations that promote cancer are sometimes rare is important for precision medicine since it would help in their identification. Driver mutations are largely discovered through their frequencies. Thus, rare mutations often escape detection. Unlike high-frequency drivers, low-frequency drivers can be tissue specific; rare drivers have extremely low frequencies. Here, we discuss rare drivers and strategies to discover them. We suggest that allosteric driver mutations shift the protein ensemble from the inactive to the active state. Rare allosteric drivers are statistically rare since, to switch the protein functional state, they cooperate with additional mutations, and these are not considered in the patient cancer-specific protein sequence analysis. A complete landscape of mutations that drive cancer will reveal tumor-specific therapeutic vulnerabilities.

A polydopamine-based platform for anti-cancer drug delivery

Cancer is the second leading cause of death in the world with around 9.6 million deaths in 2018, approximately 70% of which occurred in the middle- and low-income countries; moreover, the economic impact of cancer is significant and escalating day by day. The total annual economic cost of cancer treatment in 2010 was estimated at approximately US$ 1.16 trillion. Researchers have explored cancer mitigation therapies such as chemo-thermal therapy, chemo-photothermal therapy and photodynamic-photothermal therapy. These combinational therapies facilitate better control on the tunability of the carrier for effectively diminishing cancer cells than individual therapies such as chemotherapy, photothermal therapy and targeted therapy. All these therapies come under novel drug delivery systems in which anti-cancer drugs attack the cancerous cells due to various stimuli (e.g. pH, thermal, UV, IR, acoustic and magnetic)-responsive properties of the anti-cancer drug carriers. Compared to conventional drug delivery systems, the novel drug delivery systems have several advantages such as targeted drug release, sustained and consistent blood levels within the therapeutic window, and decreased dosing frequency. Among the numerous polymeric carriers developed for drug delivery, polydopamine has been found to be more suitable as a carrier for these drug delivery functions due to its easy and cost-effective fabrication, excellent biocompatibility, multi-drug carrier capacity and stimuli sensitivity. Therefore, in this review, we have explored polydopamine-based carriers for anti-cancer drug delivery systems to mitigate cancer and simultaneously discussed basic synthesis routes for polydopamine.

Leveraging protein dynamics to identify cancer mutational hotspots using 3D structures

Large-scale exome sequencing of tumors has enabled the identification of cancer drivers using recurrence-based approaches. Some of these methods also employ 3D protein structures to identify mutational hotspots in cancer-associated genes. In determining such mutational clusters in structures, existing approaches overlook protein dynamics, despite its essential role in protein function. We present a framework to identify cancer driver genes using a dynamics-based search of mutational hotspot communities. Mutations are mapped to protein structures, which are partitioned into distinct residue communities. These communities are identified in a framework where residue-residue contact edges are weighted by correlated motions (as inferred by dynamics-based models). We then search for signals of positive selection among these residue communities to identify putative driver genes, while applying our method to the TCGA (The Cancer Genome Atlas) PanCancer Atlas missense mutation catalog. Overall, we predict 1 or more mutational hotspots within the resolved structures of proteins encoded by 434 genes. These genes were enriched among biological processes associated with tumor progression. Additionally, a comparison between our approach and existing cancer hotspot detection methods using structural data suggests that including protein dynamics significantly increases the sensitivity of driver detection.

Integration of Random Forest Classifiers and Deep Convolutional Neural Networks for Classification and Biomolecular Modeling of Cancer Driver Mutations

Development of machine learning solutions for prediction of functional and clinical significance of cancer driver genes and mutations are paramount in modern biomedical research and have gained a significant momentum in a recent decade. In this work, we integrate different machine learning approaches, including tree based methods, random forest and gradient boosted tree (GBT) classifiers along with deep convolutional neural networks (CNN) for prediction of cancer driver mutations in the genomic datasets. The feasibility of CNN in using raw nucleotide sequences for classification of cancer driver mutations was initially explored by employing label encoding, one hot encoding, and embedding to preprocess the DNA information. These classifiers were benchmarked against their tree-based alternatives in order to evaluate the performance on a relative scale. We then integrated DNA-based scores generated by CNN with various categories of conservational, evolutionary and functional features into a generalized random forest classifier. The results of this study have demonstrated that CNN can learn high level features from genomic information that are complementary to the ensemble-based predictors often employed for classification of cancer mutations. By combining deep learning-generated score with only two main ensemble-based functional features, we can achieve a superior performance of various machine learning classifiers. Our findings have also suggested that synergy of nucleotide-based deep learning scores and integrated metrics derived from protein sequence conservation scores can allow for robust classification of cancer driver mutations with a limited number of highly informative features. Machine learning predictions are leveraged in molecular simulations, protein stability, and network-based analysis of cancer mutations in the protein kinase genes to obtain insights about molecular signatures of driver mutations and enhance the interpretability of cancer-specific classification models.