Natural product scores and fingerprints extracted from artificial neural networks

Menke, Janosch; Massa, Joana; Koch, Oliver

doi:10.1016/j.csbj.2021.07.032

Cited by 14 publications

(17 citation statements)

References 36 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The NPASS database is a curated natural product library that contains sufficient amounts of both active and inactive natural products. It has been widely used as novel drug candidates against SARS-CoV-2 [ 51 , 52 ]. From these compounds with the greatest binding affinity as determined by the molecular docking score, the top four were selected i.e., NPA000018 (Lolicine A) (−11.9 kcal/mol), NPA000111 (Armochaetoglobin N) (−11.2 kcal/mol), NPA000305 (Quartromicin D2) (−11.8 kcal/mol), and NPA014826 (Pyralomicin 2b) (−14.2 kcal/mol) ( Table 1 ).…”

Section: Discussionmentioning

confidence: 99%

High-Throughput Screening and Molecular Dynamics Simulation of Natural Products for the Identification of Anticancer Agents against MCM7 Protein

Zhang

Chen

Lin

et al. 2022

Applied Bionics and Biomechanics

View full text Add to dashboard Cite

Minichromosome maintenance complex component 7 (MCM7) belongs to the minichromosome maintenance family that is necessary for the initiation of eukaryotic DNA replication. Overexpression of the MCM7 protein is linked to cellular proliferation and is accountable for critical malignancy in many cancers. Mechanistically, the suppression of MCM7 greatly lowers the cellular proliferation associated with cancer. Advances in immunotherapy have revolutionized treatments for many types of cancer. To date, no effective small molecular candidate has been found that can stop the advancement of cancer produced by the MCM7 protein. Here, we present the findings of methods that used a combination of structure-assisted drug design, high-throughput virtual screening, and simulations studies to swiftly generate lead compounds against MCM7 protein. In the current study, we designed efficient compounds that may combat all emerging cancer targeting the common MCM7 protein. For this objective, a molecular docking and molecular dynamics (MD) simulation-based virtual screening of 29,000 NPASS library was carried out. As a consequence of using specific pharmacological, physiological, and ADMET criteria, four new prevailing compounds, NPA000018, NPA000111, NPA00305, and NPA014826, were successfully selected. The MD simulations were also used for a time period of 50 ns to evaluate for stability and dynamics behavior of the compounds. Eventually, compounds NPA000111 and NPA014826 were found to be highly potent against MCM7 protein. According to our results, the selected compounds may be effective in treating certain cancer subtypes, for which additional follow-up experimental validation is recommended.

show abstract

Section: Discussionmentioning

confidence: 99%

High-Throughput Screening and Molecular Dynamics Simulation of Natural Products for the Identification of Anticancer Agents against MCM7 Protein

Zhang

Chen

Lin

et al. 2022

Applied Bionics and Biomechanics

View full text Add to dashboard Cite

show abstract

“…Typically, the NP score falls somewhere in the range of −5 to 5. If the score is higher, then there is a greater likelihood that the molecule in question is an NP [ 23 , 24 ]. All the designed molecules displayed NP-like properties except for 30 , which showed a −9.932 NP score.…”

Section: Resultsmentioning

confidence: 99%

Fighting Antibiotic Resistance: New Pyrimidine-Clubbed Benzimidazole Derivatives as Potential DHFR Inhibitors

Haque

Marathakam²,

Rana³

et al. 2023

Molecules

View full text Add to dashboard Cite

The present work describes the design and development of seventeen pyrimidine-clubbed benzimidazole derivatives as potential dihydrofolate reductase (DHFR) inhibitors. These compounds were filtered by using ADMET, drug-likeness characteristics calculations, and molecular docking experiments. Compounds 27, 29, 30, 33, 37, 38, and 41 were chosen for the synthesis based on the results of the in silico screening. Each of the synthesized compounds was tested for its in vitro antibacterial and antifungal activities using a variety of strains. All the compounds showed antibacterial properties against Gram-positive bacteria (Staphylococcus aureus and Staphylococcus pyogenes) as well as Gram-negative bacteria (Escherichia coli and Pseudomonas aeruginosa). Most of the compounds either had a higher potency than chloramphenicol or an equivalent potency to ciprofloxacin. Compounds 29 and 33 were effective against all the bacterial and fungal strains. Finally, the 1,2,3,4-tetrahydropyrimidine-2-thiol derivatives with a 6-chloro-2-(chloromethyl)-1H-benzo[d]imidazole moiety are potent enough to be considered a promising lead for the discovery of an effective antibacterial agent.

show abstract

“…Around the value 0 are the pseudo-natural products. 30 Figure 10 presents a heat map of the values' frequency in the SA score range to compare the results achieved by each of the generated flavors. The heat map presents the range of the analyzed metrics in the y axis and the according flavor in the x axis.…”

Section: Resultsmentioning

confidence: 99%

“…The score ranged between −5 and 5; the higher the score, the more alike is the analyzed molecule to a natural product. Around the value 0 are the pseudo-natural products …”

Section: Resultsmentioning

confidence: 99%

Transfer Learning Approach to Develop Natural Molecules with Specific Flavor Requirements

Queiroz

Rebello

Costa

et al. 2023

Ind. Eng. Chem. Res.

View full text Add to dashboard Cite

In the past decades, the flavor industry’s investment in research and development has increased to take innovative steps. Meanwhile, the lack of information regarding the flavored molecules and specific flavoring properties is an obstacle to advances in this sector. In this context, this work presents the implementation of three scientific machine learning techniques as an innovative methodology to design new natural flavor molecules with specific desired properties to product development. The transfer learning technique is presented to tackle the lack of data available when analyzing flavor molecules. Nine flavor descriptors were studied along this work, and all of them presented more than 50% of molecules generated within the outstanding results considered for the evaluation metric, natural product-likeness score and synthetic accessibility score. Finally, a discussion of the results is constructed based on the data availability, the presence in nature, and the multisensorial flavor component impact for the specific flavors’ results.

show abstract

Natural product scores and fingerprints extracted from artificial neural networks

Abstract: Graphical abstract

Cited by 14 publications

References 36 publications

High-Throughput Screening and Molecular Dynamics Simulation of Natural Products for the Identification of Anticancer Agents against MCM7 Protein

High-Throughput Screening and Molecular Dynamics Simulation of Natural Products for the Identification of Anticancer Agents against MCM7 Protein

Fighting Antibiotic Resistance: New Pyrimidine-Clubbed Benzimidazole Derivatives as Potential DHFR Inhibitors

Transfer Learning Approach to Develop Natural Molecules with Specific Flavor Requirements

Contact Info

Product

Resources

About