1
|
Morales JF, Chuguransky S, Alberca LN, Alice JI, Goicoechea S, Ruiz ME, Bellera CL, Talevi A. Positive Predictive Value Surfaces as a Complementary Tool to Assess the Performance of Virtual Screening Methods. Mini Rev Med Chem 2021; 20:1447-1460. [PMID: 32072906 DOI: 10.2174/1871525718666200219130229] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 09/11/2019] [Revised: 10/28/2019] [Accepted: 10/29/2019] [Indexed: 11/22/2022]
Abstract
BACKGROUND Since their introduction in the virtual screening field, Receiver Operating Characteristic (ROC) curve-derived metrics have been widely used for benchmarking of computational methods and algorithms intended for virtual screening applications. Whereas in classification problems, the ratio between sensitivity and specificity for a given score value is very informative, a practical concern in virtual screening campaigns is to predict the actual probability that a predicted hit will prove truly active when submitted to experimental testing (in other words, the Positive Predictive Value - PPV). Estimation of such probability is however, obstructed due to its dependency on the yield of actives of the screened library, which cannot be known a priori. OBJECTIVE To explore the use of PPV surfaces derived from simulated ranking experiments (retrospective virtual screening) as a complementary tool to ROC curves, for both benchmarking and optimization of score cutoff values. METHODS The utility of the proposed approach is assessed in retrospective virtual screening experiments with four datasets used to infer QSAR classifiers: inhibitors of Trypanosoma cruzi trypanothione synthetase; inhibitors of Trypanosoma brucei N-myristoyltransferase; inhibitors of GABA transaminase and anticonvulsant activity in the 6 Hz seizure model. RESULTS Besides illustrating the utility of PPV surfaces to compare the performance of machine learning models for virtual screening applications and to select an adequate score threshold, our results also suggest that ensemble learning provides models with better predictivity and more robust behavior. CONCLUSION PPV surfaces are valuable tools to assess virtual screening tools and choose score thresholds to be applied in prospective in silico screens. Ensemble learning approaches seem to consistently lead to improved predictivity and robustness.
Collapse
Affiliation(s)
- Juan F Morales
- Laboratory of Bioactive Research and Development (LIDeB), Department of Biological Sciences, Faculty of Exact Sciences, University of La Plata (UNLP) - 47 & 115, La Plata (1900), Buenos Aires, Argentina
| | - Sara Chuguransky
- Laboratory of Bioactive Research and Development (LIDeB), Department of Biological Sciences, Faculty of Exact Sciences, University of La Plata (UNLP) - 47 & 115, La Plata (1900), Buenos Aires, Argentina
| | - Lucas N Alberca
- Laboratory of Bioactive Research and Development (LIDeB), Department of Biological Sciences, Faculty of Exact Sciences, University of La Plata (UNLP) - 47 & 115, La Plata (1900), Buenos Aires, Argentina
| | - Juan I Alice
- Laboratory of Bioactive Research and Development (LIDeB), Department of Biological Sciences, Faculty of Exact Sciences, University of La Plata (UNLP) - 47 & 115, La Plata (1900), Buenos Aires, Argentina
| | - Sofía Goicoechea
- Laboratory of Bioactive Research and Development (LIDeB), Department of Biological Sciences, Faculty of Exact Sciences, University of La Plata (UNLP) - 47 & 115, La Plata (1900), Buenos Aires, Argentina
| | - María E Ruiz
- Laboratory of Bioactive Research and Development (LIDeB), Department of Biological Sciences, Faculty of Exact Sciences, University of La Plata (UNLP) - 47 & 115, La Plata (1900), Buenos Aires, Argentina
| | - Carolina L Bellera
- Laboratory of Bioactive Research and Development (LIDeB), Department of Biological Sciences, Faculty of Exact Sciences, University of La Plata (UNLP) - 47 & 115, La Plata (1900), Buenos Aires, Argentina
| | - Alan Talevi
- Laboratory of Bioactive Research and Development (LIDeB), Department of Biological Sciences, Faculty of Exact Sciences, University of La Plata (UNLP) - 47 & 115, La Plata (1900), Buenos Aires, Argentina
| |
Collapse
|
2
|
Gunia-Krzyżak A, Żesławska E, Słoczyńska K, Koczurkiewicz P, Nitek W, Żelaszczyk D, Szkaradek N, Waszkielewicz AM, Pękala E, Marona H. Anticonvulsant activity, crystal structures, and preliminary safety evaluation of N-trans-cinnamoyl derivatives of selected (un)modified aminoalkanols. Eur J Med Chem 2015; 107:26-37. [PMID: 26560050 DOI: 10.1016/j.ejmech.2015.10.051] [Citation(s) in RCA: 14] [Impact Index Per Article: 1.6] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 07/04/2015] [Revised: 09/26/2015] [Accepted: 10/28/2015] [Indexed: 12/13/2022]
Abstract
Adequate control of seizures remains an unmet need in epilepsy. In order to identify new anticonvulsant agents, a series of N-trans-cinnamoyl derivatives of selected aminoalkanols was synthetized. The compounds were obtained in the reaction of N-acylation carried out in a two-phase system. The substances were tested in animal models of seizures induced either electrically (maximal electroshock--MES; 6-Hz test) or chemically, by subcutaneous injection of pentetrazol (scPTZ). Neurotoxicity was determined by the rotarod test. Lipophilicity of the active compounds, expressed as RM0, was determined by reversed-phase thin layer chromatography and it ranged from 1.390 to 2.219. From among the tested series of compounds, R,S-(E)-N-(1-hydroxypropan-2-yl)-3-phenylprop-2-enamide (1) and R,S-(E)-N-(2-hydroxypropyl)-3-phenylprop-2-enamide (3) exhibited the best anticonvulsant activity. Compound 1, when administered to mice by intraperitoneal (i.p.) injection, showed the ED50 values of 86.6, 60.9, and 109.6 mg/kg in the MES, 6-Hz, and scPTZ tests, respectively. For compound 3, the ED50 values were found to be 47.1 mg/kg in MES and 77.1 mg/kg in scPTZ (mice, i.p.). The distances measured in crystals of compound 1 were: 7.99 Å--from the phenyl ring to the hydroxyl group in the amide moiety, 5.729 Å--from the phenyl ring to the amide group, and 3.112 Å--from the amide group to the hydroxyl group in the amide moiety. The reported compounds did not exhibit mutagenic potential when assayed in the Ames test. Compounds 1 and 3 did not affect viability and morphology of human hepatocellular carcinoma cells (HepG2).
Collapse
Affiliation(s)
- Agnieszka Gunia-Krzyżak
- Department of Bioorganic Chemistry, Chair of Organic Chemistry, Faculty of Pharmacy, Jagiellonian University Medical College, Medyczna 9, 30-688 Kraków, Poland.
| | - Ewa Żesławska
- Departament of Chemistry, Institute of Biology, Pedagogical University, Podchorążych 2, 30-084 Kraków, Poland
| | - Karolina Słoczyńska
- Department of Pharmaceutical Biochemistry, Faculty of Pharmacy, Jagiellonian University Medical College, Medyczna 9, 30-688 Kraków, Poland
| | - Paulina Koczurkiewicz
- Department of Pharmaceutical Biochemistry, Faculty of Pharmacy, Jagiellonian University Medical College, Medyczna 9, 30-688 Kraków, Poland
| | - Wojciech Nitek
- Faculty of Chemistry, Jagiellonian University, Ingardena 3, 30-060 Kraków, Poland
| | - Dorota Żelaszczyk
- Department of Bioorganic Chemistry, Chair of Organic Chemistry, Faculty of Pharmacy, Jagiellonian University Medical College, Medyczna 9, 30-688 Kraków, Poland
| | - Natalia Szkaradek
- Department of Bioorganic Chemistry, Chair of Organic Chemistry, Faculty of Pharmacy, Jagiellonian University Medical College, Medyczna 9, 30-688 Kraków, Poland
| | - Anna M Waszkielewicz
- Department of Bioorganic Chemistry, Chair of Organic Chemistry, Faculty of Pharmacy, Jagiellonian University Medical College, Medyczna 9, 30-688 Kraków, Poland
| | - Elżbieta Pękala
- Department of Pharmaceutical Biochemistry, Faculty of Pharmacy, Jagiellonian University Medical College, Medyczna 9, 30-688 Kraków, Poland
| | - Henryk Marona
- Department of Bioorganic Chemistry, Chair of Organic Chemistry, Faculty of Pharmacy, Jagiellonian University Medical College, Medyczna 9, 30-688 Kraków, Poland
| |
Collapse
|