A drug discovery platform to identify compounds that inhibit EGFR triple mutants

The diversification of methods for studying cell-cell interactions and communication

No cell lives in a vacuum, and the molecular interactions between cells define most phenotypes. Transcriptomics provides rich information to infer cell-cell interactions and communication, thus accelerating the discovery of the roles of cells within their communities. Such research relies heavily on algorithms that infer which cells are interacting and the ligands and receptors involved. Specific pressures on different research niches are driving the evolution of next-generation computational tools, enabling new conceptual opportunities and technological advances. More sophisticated algorithms now account for the heterogeneity and spatial organization of cells, multiple ligand types and intracellular signalling events, and enable the use of larger and more complex datasets, including single-cell and spatial transcriptomics. Similarly, new high-throughput experimental methods are increasing the number and resolution of interactions that can be analysed simultaneously. Here, we explore recent progress in cell-cell interaction research and highlight the diversification of the next generation of tools, which have yielded a rich ecosystem of tools for different applications and are enabling invaluable discoveries.

Predicting the functional state of protein kinases using interpretable graph neural networks from sequence and structural data

Protein kinases are central to cellular activities and are actively pursued as drug targets for several conditions including cancer and autoimmune diseases. Despite the availability of a large structural database for kinases, methodologies to elucidate the structure-function relationship of these proteins (without manual intervention) are lacking. Such techniques are essential in structural biology and to accelerate drug discovery efforts. Here, we implement an interpretable graph neural network (GNN) framework for classifying the functionally active and inactive states of a large set of protein kinases by only using their tertiary structure and amino acid sequence. We show that the GNN models can classify kinase structures with high accuracy (>97%). We implement the Gradient-weighted Class Activation Mapping for graphs (Graph Grad-CAM) to automatically identify structurally important residues and residue-residue contacts of the kinases without any a priori input. We show that the motifs identified through the Graph Grad-CAM methodology are functionally critical, consistent with the existing kinase literature. Notably, the highly conserved DFG and HRD motifs of the well-known hydrophobic spine are identified by the interpretable framework in addition to some of the lesser known motifs. Further, using Grad-CAM maps as the vector embedding of the protein structures, we identify the subtle differences in the crystal structures among different sub-classes of kinases in the Protein Data Bank (PDB). Frameworks such as the one implemented here, for high-throughput identification of protein structure-function relationships are essential in designing targeted small molecules therapies as well as in engineering new proteins for novel applications.

Two Main Cancer Biomarkers as Molecular Targets of Binase Antitumor Activity

Cancer is frequently coupled with the disturbance of key signaling pathways. Aberrant activation of the mitogen-activated protein kinase (MAPK) signaling cascade, occurring in over 85% of cancers, is mainly caused by the genetic alterations of its main components-oncogenes EGFR and RAS, and plays a crucial role in cell fate. The importance of EGFR and RAS proteins in a variety of tumors suggests that they would be good therapeutic targets, but at present, no effective targeted therapy against these two oncogenes has been proven. Here, we show that ribonuclease from Bacillus pumilus (binase) inhibits MAPK signaling through direct interaction with EGFR and RAS proteins. This effect contributes to the antitumor potential of binase along with its enzymatic activity. Multitargeticity of binase prevents the development of drug resistance, which is considered a major obstacle to effective anticancer treatment.

Programmable synthetic receptors: the next-generation of cell and gene therapies

Cell and gene therapies hold tremendous promise for treating a range of difficult-to-treat diseases. However, concerns over the safety and efficacy require to be further addressed in order to realize their full potential. Synthetic receptors, a synthetic biology tool that can precisely control the function of therapeutic cells and genetic modules, have been rapidly developed and applied as a powerful solution. Delicately designed and engineered, they can be applied to finetune the therapeutic activities, i.e., to regulate production of dosed, bioactive payloads by sensing and processing user-defined signals or biomarkers. This review provides an overview of diverse synthetic receptor systems being used to reprogram therapeutic cells and their wide applications in biomedical research. With a special focus on four synthetic receptor systems at the forefront, including chimeric antigen receptors (CARs) and synthetic Notch (synNotch) receptors, we address the generalized strategies to design, construct and improve synthetic receptors. Meanwhile, we also highlight the expanding landscape of therapeutic applications of the synthetic receptor systems as well as current challenges in their clinical translation.

MMD collaborates with ACSL4 and MBOAT7 to promote polyunsaturated phosphatidylinositol remodeling and susceptibility to ferroptosis

Ferroptosis is a form of regulated cell death with roles in degenerative diseases and cancer. Excessive iron-catalyzed peroxidation of membrane phospholipids, especially those containing the polyunsaturated fatty acid arachidonic acid (AA), is central in driving ferroptosis. Here, we reveal that an understudied Golgi-resident scaffold protein, MMD, promotes susceptibility to ferroptosis in ovarian and renal carcinoma cells in an ACSL4- and MBOAT7-dependent manner. Mechanistically, MMD physically interacts with both ACSL4 and MBOAT7, two enzymes that catalyze sequential steps to incorporate AA in phosphatidylinositol (PI) lipids. Thus, MMD increases the flux of AA into PI, resulting in heightened cellular levels of AA-PI and other AA-containing phospholipid species. This molecular mechanism points to a pro-ferroptotic role for MBOAT7 and AA-PI, with potential therapeutic implications, and reveals that MMD is an important regulator of cellular lipid metabolism.

Integrated and High-Throughput Approach for Sensitive Analysis of Tyrosine Phosphoproteome

Tyrosine phosphorylation (pTyr) regulates various signaling pathways under normal and cancerous states. Due to their low abundance and transient and dynamic natures, systematic profiling of pTyr sites is challenging. Antibody and engineered binding domain-based approaches have been well applied to pTyr peptide enrichment. However, traditional methods have the disadvantage of a long sample preparation process, which makes them unsuitable for processing limited amount of samples, especially in a high-throughput manner. In this study we developed a 96-well microplate-based approach to integrate all the sample preparation steps starting from cell culture to MS-compatible pTyr peptide enrichment in three consecutive 96-well microplates. By assembling an engineered SH2 domain onto a microplate, nonspecific adsorption of phosphopeptides is greatly reduced, which allows us to remove the Ti-IMAC purification and three C18 desalting steps (after digestion, pTyr enrichment, and Ti-IMAC purification) and, therefore, greatly simplifies the entire pTyr peptide enrichment workflow, especially when processing a large number of samples. Starting with 96-well microplate-cultured, pervanadate-stimulated cells, our approach could enrich 21% more pTyr sites than the traditional serial pTyr enrichment approach and showed good sensitivity and reproducibility in the range of 200 ng to 200 μg peptides. Importantly, we applied this approach to profile tyrosine kinase inhibitor-mediated EGFR signaling pathway and could well differentiate the distinct response of different pTyr sites. Collectively, the integrated 96-well microplate-based approach is valuable for profiling pTyr sites from limited biological samples and in a high-throughput manner.

Prognostic value of receptor tyrosine kinases in malignant melanoma patients: A systematic review and meta-analysis of immunohistochemistry

Background

Substantial evidence suggests that receptor tyrosine kinases (RTKs) are overexpressed in tumors; however, few studies have focused on the prognostic value of RTKs in melanoma.

Objectives

The objective of this study is to evaluate the association between overexpression of RTKs and survival in melanoma patients based on immunohistochemistry (IHC) analysis.

Methods

Our review is registered on PROSPERO (http://www.crd.york.ac.uk/PROSPERO), registration number CRD42021261460. Seven databases were searched, and data were extracted. We used IHC to measure the association between overexpression of RTKs and overall survival (OS), disease-free survival (DFS), progression-free survival (PFS), and clinicopathology in melanoma patients. Pooled analysis was conducted to assess the differences between Hazard Ratios along with 95% confidence intervals.

Results

Of 5,508 publications examined following the database search, 23 publications were included in this study, which included data from a total of 2,072 patients. Vascular endothelial growth factor receptor 2 (VEGF-R2) overexpression was associated with worse OS and DFS in melanoma. Furthermore, there was an association between OS and the expression of several RTKs, including epidermal growth factor receptor (EGFR), mesenchymal-epithelial transition factor (MET), vascular endothelial growth factor receptor 1 (VEGF-R1), and insulin-like growth factor 1 receptor (IGF-1R). There were no significant correlations between EGFR overexpression and worse DFS or PFS. EGFR overexpression was associated with worse OS cutaneous and nasal melanoma, but not uveal melanoma. However, MET overexpression was related to worse OS in both cutaneous and uveal melanoma. Furthermore, EGFR overexpression was associated with a worse OS in Europe compared to other geographic areas. Moreover, EGFR and MET overexpression showed significant prognostic value in patients with the cut-off “≥10% staining”.

Conclusions

Our findings build concrete evidence that overexpression of RTKs is associated with poor prognosis and clinicopathology in melanoma, highlighting RTK expression has the potential to inform individualized combination therapies and accurate prognostic evaluation.

Indexing Ultrafast Shape-Based Descriptors in MongoDB to Identify TLR4 Pathway Agonists

A method is presented for an ultrafast shape-based search workflow for the screening of large compound collections, i.e., those of vendors. The three-dimensional shape of a molecule dictates its biological activity by enabling the molecule to fit into binding pockets of proteins. Quite often, distinctly different chemical compounds that have similar shapes can bind in a similar way. OpenEye pioneered an algorithm for comparing shapes of molecules by overlaying them in a computer and measuring differences between a query molecule and a target molecule. Overlaying shapes is a computationally intensive process and represents a bottleneck in searching for similar molecules. More recent publications describe alternative methods of overlaying molecules, which are accomplished by comparing shape-based descriptors. These methods were implemented in the Open Drug Discovery Toolkit (ODDT) package. We utilized a combination of open-source software packages like ODDT and RDkit to implement a workflow for ultrafast conformer generation and matching that does not require storing precomputed conformers on the file system or in memory. Moreover, the generated descriptors could be optionally stored in MongoDB for performing searches in the future. To speed up the search, we created a set of indexes from the transformed shape-based descriptors. We are in the process of calculating descriptors for multiple vendors, including Enamine's "REAL" collection of 1.2 billion compounds. Currently, the shape similarity search on more than 70 million compounds takes less than 8 s! We exemplified our methodology with the screen of compounds that can act as putative TLR4 agonists. The search was based on a literature-known small-molecule TLR4 agonist series. In due course, we identified compounds with novel structural motifs that were active in mouse and human TLR4 reporter cell lines.

CFTR interactome mapping using the mammalian membrane two‐hybrid high‐throughput screening system

Cystic Fibrosis Transmembrane Conductance Regulator (CFTR) is a chloride and bicarbonate channel in secretory epithelia with a critical role in maintaining fluid homeostasis. Mutations in CFTR are associated with Cystic Fibrosis (CF), the most common lethal autosomal recessive disorder in Caucasians. While remarkable treatment advances have been made recently in the form of modulator drugs directly rescuing CFTR dysfunction, there is still considerable scope for improvement of therapeutic effectiveness. Here, we report the application of a high‐throughput screening variant of the Mammalian Membrane Two‐Hybrid (MaMTH‐HTS) to map the protein–protein interactions of wild‐type (wt) and mutant CFTR (F508del), in an effort to better understand CF cellular effects and identify new drug targets for patient‐specific treatments. Combined with functional validation in multiple disease models, we have uncovered candidate proteins with potential roles in CFTR function/CF pathophysiology, including Fibrinogen Like 2 (FGL2), which we demonstrate in patient‐derived intestinal organoids has a significant effect on CFTR functional expression.

TKI resistant-based prognostic immune related gene signature in LUAD, in which FSCN1 contributes to tumor progression

Drug resistance reflects the evolution of tumors, which is the main cause of recurrence and death. Currently, EGFR-TKI treatment is the first-line therapy for lung adenocarcinoma (LUAD) patients. Although EGFR-TKI achieved good effects at the beginning, most of the LUAD patients eventually acquired resistance. Therefore, it's urgently need to develop a strong criterion for identifying these patients who may benefit from additional therapy. In this study, we established a three TKI resistant-related gene signature (DDIT4, OAS3, FSCN1), and determined that's an accuracy, independent and specific prognostic model for LUAD patients. Patients categorized as high-risk by this signature showed more sensitive to chemotherapy, and exhibited higher expression of common immune checkpoints such as PD-L1/B3H7/PD-L2/IDO1. Moreover, these patients were characterized by increased infiltration of M0 macrophage and activated memory CD4⁺ T cells. The expression and prognostic values of DDIT4, FSCN1 and OAS3 were further confirmed in clinical data. In addition, experimental data showed that FSCN1 promoted LUAD development via PI3K/AKT signaling. In conclusion, this signature is highly predictive of prognostic in LUAD patients, and may serve as a powerful prediction tool for LUAD patients to further choose chemo- and immunotherapies.

D154Q Mutation does not Alter KRAS Dimerization

KRAS is one of the most frequently mutated oncogenes in human cancers. Despite nearly 40 years of research, KRAS remains largely undruggable, in part due to an incomplete understanding of its biology. Recently, KRAS dimerization was discovered to play an important role in its signalling function. The KRAS D154Q mutant was described as a dimer-deficient variant that can be used to study the effect of dimerization in KRAS oncogenicity. However, we show here that KRAS D154Q homo- and heterodimerized with KRAS WT using three separate protein-protein interaction assays, and that oncogenic KRAS dimerization was not negatively impacted by the presence of a secondary D154Q mutation. In conclusion, we advise caution in using this variant to study the purpose of dimerization in KRAS oncogenic behaviour.

Chemical Genetics Screen Identifies COPB2 Tool Compounds That Alters ER Stress Response and Induces RTK Dysregulation in Lung Cancer Cells

Activating mutations in the epidermal growth factor receptor (EGFR) are common driver mutations in non-small cell lung cancer (NSCLC). First, second and third generation EGFR tyrosine kinase inhibitors (TKIs) are effective at inhibiting mutant EGFR NSCLC, however, acquired resistance is a major issue, leading to disease relapse. Here, we characterize a small molecule, EMI66, an analog of a small molecule which we previously identified to inhibit mutant EGFR signalling via a novel mechanism of action. We show that EMI66 attenuates receptor tyrosine kinase (RTK) expression and signalling and alters the electrophoretic mobility of Coatomer Protein Complex Beta 2 (COPB2) protein in mutant EGFR NSCLC cells. Moreover, we demonstrate that EMI66 can alter the subcellular localization of EGFR and COPB2 within the early secretory pathway. Furthermore, we find that COPB2 knockdown reduces the growth of mutant EGFR lung cancer cells, alters the post-translational processing of RTKs, and alters the endoplasmic reticulum (ER) stress response pathway. Lastly, we show that EMI66 treatment also alters the ER stress response pathway and inhibits the growth of mutant EGFR lung cancer cells and organoids. Our results demonstrate that targeting of COPB2 with EMI66 presents a viable approach to attenuate mutant EGFR signalling and growth in NSCLC.

Mapping the Phospho-dependent ALK Interactome to Identify Novel Components in ALK Signaling

Protein-protein interactions (PPIs) play essential roles in Anaplastic Lymphoma Kinase (ALK) signaling. Systematic characterization of ALK interactors helps elucidate novel ALK signaling mechanisms and may aid in the identification of novel therapeutics targeting related diseases. In this study, we used the Mammalian Membrane Two-Hybrid (MaMTH) system to map the phospho-dependent ALK interactome. By screening a library of 86 SH2 domain-containing full length proteins, 30 novel ALK interactors were identified. Many of their interactions are correlated to ALK phosphorylation activity: oncogenic ALK mutations potentiate the interactions and ALK inhibitors attenuate the interactions. Among the novel interactors, NCK2 was further verified in neuroblastoma cells using co-immunoprecipitation. Modulation of ALK activity by addition of inhibitors lead to concomitant changes in the tyrosine phosphorylation status of NCK2 in neuroblastoma cells, strongly supporting the functionality of the ALK/NCK2 interaction. Our study provides a resource list of potential novel ALK signaling components for further study.

Dissecting intercellular and intracellular signaling networks with barcoded genetic tools

The power of next-generation sequencing has stimulated the development of many analysis techniques for transcriptomics and genomics. More recently, the concept of 'molecular barcoding' has broadened the spectrum of sequencing-based applications to dissect different aspects of intracellular and intercellular signaling. In these assay formats, barcode reporters replace standard reporter genes. The virtually infinitive number of expressed barcode sequences allows high levels of multiplexing, hence accelerating experimental progress. Furthermore, reporter barcodes are used to quantitatively monitor a variety of biological events in living cells which has already provided much insight into complex cellular signaling and will further increase our knowledge in the future.

Drugging the undruggable proteins in cancer: A systems biology approach

In recent years, the research community has, with comprehensive systems biology approaches and related technologies, gained insight into the vast complexity of numerous cancers. These approaches allow an in-depth exploration that cannot be achieved solely using conventional low-throughput methods, which do not closely mimic the natural cellular environment. In this review, we discuss recent integrative multiple omics approaches for understanding and modulating previously identified 'undruggable' targets such as members of the RAS family, MYC, TP53, and various E3 ligases and deubiquitinases. We describe how these technologies have revolutionized drug discovery by overcoming an array of biological and technological challenges and how, in the future, they will be pivotal in assessing cancer states in individual patients, allowing for the prediction and application of personalized disease treatments.

Synthetic gene circuits as tools for drug discovery

Within mammalian systems, there exists enormous opportunity to use synthetic gene circuits to enhance phenotype-based drug discovery, to map the molecular origins of disease, and to validate therapeutics in complex cellular systems. While drug discovery has relied on marker staining and high-content imaging in cell-based assays, synthetic gene circuits expand the potential for precision and speed. Here we present a vision of how circuits can improve the speed and accuracy of drug discovery by enhancing the efficiency of hit triage, capturing disease-relevant dynamics in cell-based assays, and simplifying validation and readouts from organoids and microphysiological systems (MPS). By tracking events and cellular states across multiple length and time scales, circuits will transform how we decipher the causal link between molecular events and phenotypes to improve the selectivity and sensitivity of cell-based assays.

Receptor tyrosine kinases and cancer: oncogenic mechanisms and therapeutic approaches

Receptor tyrosine kinases (RTKs) are transmembrane receptors of great clinical interest due to their role in disease, notably cancer. Since their discovery, several mechanisms of RTK dysregulation have been identified, resulting in multiple cancer types displaying 'oncogenic addiction' to RTKs. As a result, RTKs have represented a major class for targeted therapeutics over the past two decades, with numerous small molecule-based tyrosine kinase inhibitor (TKI) therapeutics having been developed and clinically approved for several cancers. However, many of the current RTK inhibitor treatments eventually result in the rapid development of acquired resistance and subsequent tumor relapse. Recent technological advances and tools are being generated for the identification of novel RTK small molecule therapeutics. These newer technologies will be important for the identification of diverse types of RTK inhibitors, targeting both the receptors themselves as well as key cellular factors that play important roles in the RTK signaling cascade.

A homogeneous split-luciferase assay for rapid and sensitive detection of anti-SARS CoV-2 antibodies

Better diagnostic tools are needed to combat the ongoing COVID-19 pandemic. Here, to meet this urgent demand, we report a homogeneous immunoassay to detect IgG antibodies against SARS-CoV-2. This serological assay, called SATiN, is based on a tri-part Nanoluciferase (tNLuc) approach, in which the spike protein of SARS-CoV-2 and protein G, fused respectively to two different tNLuc tags, are used as antibody probes. Target engagement of the probes allows reconstitution of a functional luciferase in the presence of the third tNLuc component. The assay is performed directly in the liquid phase of patient sera and enables rapid, quantitative and low-cost detection. We show that SATiN has a similar sensitivity to ELISA, and its readouts are consistent with various neutralizing antibody assays. This proof-of-principle study suggests potential applications in diagnostics, as well as disease and vaccination management.

Tackling Drug Resistance in EGFR Exon 20 Insertion Mutant Lung Cancer

Insertion mutations in exon 20 (Ex20ins) of the epidermal growth factor receptor (EGFR) gene are the largest class of EGFR mutations in non-small cell lung cancer (NSCLC) for which there are currently no approved targeted therapies. NSCLC patients with these mutations do not respond to clinically approved EGFR tyrosine kinase inhibitors (TKIs) and have poor outcomes. A number of early phase clinical trials are currently underway to evaluate the efficacy of a new generation of TKIs that are capable of binding to and blocking Ex20ins. Although these agents have shown some clinical activity, patient responses have been restricted by dose-limiting toxicity or rapid acquisition of resistance after a short response. Here we review the current understanding of the mechanisms of resistance to these compounds, which include on-target EGFR secondary mutations, compensatory bypass pathway activation and acquisition of an EMT phenotype. Taking lessons from conventional EGFR inhibitor therapy in NSCLC, we also consider other potential sources of resistance including the presence of drug-tolerant persister cells. We will discuss therapeutic strategies which have the potential to overcome different forms of drug resistance. We conclude by evaluating recent technological developments in drug discovery such as PROTACs as a means to better tackle TKI resistance in NSCLC harbouring Ex20ins mutations.

Developing drugs for the 'undruggable'

There are up to 650,000 'undruggable' protein-protein interactions (PPIs) in the human interactome that can be potentially considered as novel therapeutic targets. How does the 'undruggable' become 'druggable'?

Mitochondria under the spotlight: On the implications of mitochondrial dysfunction and its connectivity to neuropsychiatric disorders

Neuropsychiatric disorders (NPDs) such as bipolar disorder (BD), schizophrenia (SZ) and mood disorder (MD) are hard to manage due to overlapping symptoms and lack of biomarkers. Risk alleles of BD/SZ/MD are emerging, with evidence suggesting mitochondrial (mt) dysfunction as a critical factor for disease onset and progression. Mood stabilizing treatments for these disorders are scarce, revealing the need for biomarker discovery and artificial intelligence approaches to design synthetically accessible novel therapeutics. Here, we show mt involvement in NPDs by associating 245 mt proteins to BD/SZ/MD, with 7 common players in these disease categories. Analysis of over 650 publications suggests that 245 NPD-linked mt proteins are associated with 800 other mt proteins, with mt impairment likely to rewire these interactions. High dosage of mood stabilizers is known to alleviate manic episodes, but which compounds target mt pathways is another gap in the field that we address through mood stabilizer-gene interaction analysis of 37 prescriptions and over-the-counter psychotropic treatments, which we have refined to 15 mood-stabilizing agents. We show 26 of the 245 NPD-linked mt proteins are uniquely or commonly targeted by one or more of these mood stabilizers. Further, induced pluripotent stem cell-derived patient neurons and three-dimensional human brain organoids as reliable BD/SZ/MD models are outlined, along with multiomics methods and machine learning-based decision making tools for biomarker discovery, which remains a bottleneck for precision psychiatry medicine.

Structure-guided discovery of a luminescent theranostic toolkit for living cancer cells and the imaging behavior effect

Dual-functional theranostics are powerful tools that can allow for the in-field understanding of cancer pathology, yet their use is held back by the paucity of suitable theranostics for living systems. Moreover, typical in vitro screening conditions for probe molecules do not necessarily generate candidates that can function effectively in the natural in cellulo environment, limiting their follow-up use in living systems. We introduce herein a general strategy for the development of an iridium(iii) theranostic by grafting a well-known inhibitor as a "binding unit" onto an iridium(iii) complex precursor as a "signaling unit". To further optimize their emissive properties, we explored the effect of imaging behavior by incorporating different substituents onto the parental "signaling unit". This design concept was validated by a series of tailored iridium(iii) theranostics 2a-2h for the visualization and inhibition of EGFR in living cancer cells. By comprehensively assessing the theranostic potency of 2a-2h in both in vitro and in cellulo contexts, probe 2f containing electron-donating methoxy groups on the "signaling unit" was discovered to be the most promising candidate theranostic with desirable photophysical/chemical properties. Probe 2f selectively bound to EGFR in vitro and in cellulo, enabling it to selectively discriminate living EGFR-overexpressing cancer cells from normal cells that express low levels of EGFR with an "always-on" luminescence signal output. In particular, its long-lived lifetime enabled its luminescence signal to be readily distinguished from the interfering fluorescence of organic dyes by using time-resolved techniques. Complex 2f simultaneously visualized and inhibited EGFR in a dose-dependent manner, leading to a reduction in the phosphorylation of downstream proteins ERK and MEK, and inhibition of the activity of downstream transcription factor AP1. Notably, complex 2f is comparable to the parental EGFR inhibitor 1b, in terms of both inhibitory activity against EGFR and cytotoxicity against EGFR-overexpressing cancer cells. This tailored dual-functional iridium(iii) theranostic toolkit provides an alternative strategy for the personalized diagnosis and treatment of cancers.

DeepH-DTA: Deep Learning for Predicting Drug-Target Interactions: A Case Study of COVID-19 Drug Repurposing

The rapid spread of novel coronavirus pneumonia (COVID-19) has led to a dramatically increased mortality rate worldwide. Despite many efforts, the rapid development of an effective vaccine for this novel virus will take considerable time and relies on the identification of drug-target (DT) interactions utilizing commercially available medication to identify potential inhibitors. Motivated by this, we propose a new framework, called DeepH-DTA, for predicting DT binding affinities for heterogeneous drugs. We propose a heterogeneous graph attention (HGAT) model to learn topological information of compound molecules and bidirectional ConvLSTM layers for modeling spatio-sequential information in simplified molecular-input line-entry system (SMILES) sequences of drug data. For protein sequences, we propose a squeezed-excited dense convolutional network for learning hidden representations within amino acid sequences; while utilizing advanced embedding techniques for encoding both kinds of input sequences. The performance of DeepH-DTA is evaluated through extensive experiments against cutting-edge approaches utilising two public datasets (Davis, and KIBA) which comprise eclectic samples of the kinase protein family and the pertinent inhibitors. DeepH-DTA attains the highest Concordance Index (CI) of 0.924 and 0.927 and also achieved a mean square error (MSE) of 0.195 and 0.111 on the Davis and KIBA datasets respectively. Moreover, a study using FDA-approved drugs from the Drug Bank database is performed using DeepH-DTA to predict the affinity scores of drugs against SARS-CoV-2 amino acid sequences, and the results show that that the model can predict some of the SARS-Cov-2 inhibitors that have been recently approved in many clinical studies.