In this paper we review the current status of high-performance computing applications in the general area of drug discovery. We provide an introduction to the methodologies applied at atomic and molecular scales, followed by three specific examples of implementation of these tools. The first example describes in silico modeling of the adsorption of small molecules to organic and inorganic surfaces, which may be applied to drug delivery issues. The second example involves DNA translocation through nanopores with major significance to DNA sequencing efforts. The final example offers an overview of computer-aided drug design, with some illustrative examples of its usefulness.

Colchicine is a well-known compound with strong antiproliferative activity that has had limited use in chemotherapy because of its toxicity. In order to create more potent anticancer agents, a series of novel colchicine derivatives have been obtained by simultaneous modification at C7 (amides and sulfonamides) and at C10 (methylamino group) positions and characterized by spectroscopic methods. All the synthesized compounds have been tested in vitro to evaluate their cytotoxicity toward A549, MCF-7, LoVo, LoVo/DX and BALB/3T3 cell lines. Additionally, the activity of the studied compounds was investigated using computational methods involving molecular docking of the colchicine derivatives to β-tubulin. The majority of the obtained derivatives exhibited higher cytotoxicity than colchicine, doxorubicin or cisplatin against tested cancer cell lines. Furthermore, molecular modeling studies of the obtained compounds revealed their possible binding modes into the colchicine binding site of tubulin.

The SARS-CoV-2 virus invades and replicates within host cells by “hijacking” biomolecular machinery, gaining control of the microtubule cytoskeleton. After attaching to membrane receptors and entering cells, the SARS-CoV-2 virus co-opts the dynamic intra-cellular cytoskeletal network of microtubules, actin, and the microtubule-organizing center, enabling three factors that lead to clinical pathology: (1) viral load due to intra-cellular trafficking, (2) cell-to-cell spread by filopodia, and (3) immune dysfunction, ranging from hyper-inflammatory cytokine storm to ineffective or absent response. These factors all depend directly on microtubules and the microtubule-organizing center, as do cell functions such as mitosis and immune cell movement. Here we consider how the SARS-CoV-2 virus may “hijack” cytoskeletal functions by docking inside the microtubule-organizing center’s centriole “barrels”, enabling certain interactions between the virus’s positively charged spike (“S”) proteins and negatively charged C-termini of the microtubules that the centriole comprises, somewhat like fingers on a keyboard. This points to the potential benefit of therapies aimed not directly at the virus but at the microtubules and microtubule-organizing center of the host cell on which the virus depends. These therapies could range from anti-microtubule drugs to low-intensity ultrasound (megahertz mechanical vibrations) externally applied to the vagus nerve at the neck and/or to the spleen (since both are involved in mediating inflammatory response). Given that ultrasound imaging machines suitable for vagal/splenic ultrasound are available for clinical trials in every hospital, we recommend an alternative therapeutic approach for COVID-19 based on addressing and normalizing the host cell microtubules and microtubule-organizing centers co-opted by the SARS-CoV-2 virus.

Colchicine is a well-known anticancer compound showing antimitotic effect on cells. Its high cytotoxic activity against different cancer cell lines has been demonstrated many times. In this paper we report the syntheses and spectroscopic analyses of novel colchicine derivatives obtained by structural modifications at C7 (carbon-nitrogen single bond) and C10 (methylamino group) positions. All the obtained compounds have been tested in vitro to determine their cytotoxicity toward A549, MCF-7, LoVo, LoVo/DX, and BALB/3T3 cell lines. The majority of obtained derivatives exhibited higher cytotoxicity than colchicine, doxorubicin and cisplatin against the tested cancerous cell lines. Additionally, most of the presented derivatives were able to overcome the resistance of LoVo/DX cells. Additionally, their mode of binding to β-tubulin was evaluated in silico. Molecular docking studies showed that apart from the initial amides 1 and 2, compound 14, which had the best antiproliferative activity (IC₅₀ = 0.1–1.6 nM), stood out also in terms of its predicted binding energy and probably binds best into the active site of βI-tubulin isotype.

Background: The emergence and rapid spread of SARS-CoV-2 (severe acute respiratory syndrome coronavirus 2) in thelate 2019 has caused a devastating global pandemic of the severe pneumonia-like disease coronavirus disease 2019 (COVID-19). Although vaccines have been and are being developed, they are not accessible to everyone and not everyone can receive these vaccines. Also, it typically takes more than 10 years until a new therapeutic agent is approved for usage. Therefore, repurposing of known drugs can lend itself well as a key approach for significantly expediting the development of new therapies for COVID-19.Methods: We have incorporated machine learning-based computational tools and in silico models into the drug discovery process to predict Adsorption, Distribution, Metabolism, Excretion, and Toxicity (ADMET) profiles of 90 potential drugs for COVID-19 treatment identified from two independent studies mainly with the purpose of mitigating late-phase failures because of inferior pharmacokinetics and toxicity.Results: Here, we summarize the cardiotoxicity and general toxicity profiles of 90 potential drugs for COVID-19 treatment and outline the risks of repurposing and propose a stratification of patients accordingly. We shortlist a total of five compounds based on their non-toxic properties.Conclusion: In summary, this manuscript aims to provide a potentially useful source of essential knowledge on toxicity assessment of 90 compounds for healthcare practitioners and researchers to find off-label alternatives for the treatment for COVID-19. The majority of the molecules discussed in this manuscript have already moved into clinical trials and thus their known pharmacological and human safety profiles are expected to facilitate a fast track preclinical and clinical assessment for treating COVID-19.

The cutting-edge technique of Proteolysis Targeting Chimeras, or PROTACs, has gained significant attention as a viable approach for specific protein degradation. This innovative technology has vast potential in fields such as cancer therapy and drug development. The development of effective and specific therapies for a range of diseases is within reach with PROTACs, which can target previously "undruggable" proteins while circumventing the off-target effects of conventional small molecule inhibitors. This manuscript aims to discuss the application of in silico techniques to the design of these groundbreaking molecules and develop PROTAC complexes, in order to identify potential PROTAC candidates with favorable drug-like properties. Additionally, this manuscript reviews the strengths and weaknesses of these methods to demonstrate their utility and highlights the challenges and future prospects of in silico PROTAC design. The present review provides a valuable and beginner-friendly resource for researchers and drug developers interested in using in silico methods for PROTAC design, specifically ternary structure prediction.