This review examines the emerging role of machine learning (ML) in pharmaceutical chemistry, with emphasis on molecular design, synthetic feasibility, and structure–property–performance (SPP) relationships. By enabling pre-synthesis prediction of physicochemical properties, reaction pathways, and pharmaceutical performance, ML can reduce empirical trial-and-error experimentation and support more efficient exploration of chemical space. A structured narrative review design with PRISMA-aligned systematic search elements was used to evaluate 101 studies, enabling transparent literature identification, eligibility screening, and thematic synthesis across heterogeneous ML applications in pharmaceutical chemistry. This review examines structure–property relationships (SPRs) and property–performance relationships (PPRs), with emphasis on key pharmaceutical endpoints such as solubility, permeability, stability, dissolution, and bioavailability. An integrated SPP framework is proposed to connect molecular structure, intermediate properties, and final performance outcomes while incorporating retrosynthetic analysis and experimental feedback and closed-loop optimization. Recent frontier developments are also discussed, including molecular foundation models, multimodal language–graph models, diffusion-based molecular generation, E(3)-equivariant models, and MolMIM-like latent-space optimization. This review also covers co-folding and joint ligand–protein modeling, Boltz-2-like affinity prediction, AlphaFold 3-related biomolecular interaction modeling, and absorption, distribution, metabolism, excretion, and toxicity (ADMET) prediction. Key limitations include dataset leakage, benchmark inconsistency, assay variability, conformational and protonation-state effects, reproducibility challenges, regulatory constraints, and the gap between computational prediction and prospective experimental validation. Future progress is expected to depend on hybrid physics–ML models, uncertainty-aware prospective validation, autonomous experimentation, explainable artificial intelligence, and sustainability-aware molecular design. Overall, ML is evolving from a predictive tool into a chemically informed decision-support framework for rational, synthesis-aware, and experimentally validated pharmaceutical development.

The genus of wild rose, or Rosa L., belongs to the family Rosaceae. It has many cultural forms, bred under the general name Rosa. At the moment, more than 400 species of rose hips are known. Usually these are upright shrubs, less often creepers, sometimes low tree-like forms or almost herbaceous plants. Rosehip is common in the temperate and subtropical zones of the Northern Hemisphere, it can often be found in the mountainous regions of the tropical zone. Some species of rose hips are common from the Arctic Circle in the north to Ethiopia in the south. On the American continent - from Canada to Mexico, Favorable conditions for Rosehip are in the Mediterranean region. Several species of the Rosehip genus have an extensive distribution area. Kazakhstan has significant resources of medicinal raw materials of plant origin. Among the most priority representatives of the wild flora are species of the rosehip genus - Rosa L. as a most valuable medicinal plant. In total, 21 species of wild rose grow in the republic, including 5 in the central Kazakhstan: R. glabrifolia - sh. naked, R. laxa Retz. - w. loose, R. acicularis Lindl. - (W. needlefish), R. majalis Herrm. (R. cinnamomea L.) - sh. May (W. cinnamon) and R. pimpinellifolia L. (R. spinosissima L.) - W. femoral cell. Kazakhstan species of the genus Rosa L. Rose hips are characterized by a high content of biologically active substances and are widely used both in medicine and in cooking. In this paper, a comparative analysis of rose hips, Rosa canina L., harvested in the fall of 2018 is carried out. Plant raw materials are collected in the southern region of the Republic of Kazakhstan (Almaty region). The crushed air-dry raw materials were extracted with hexane and chloroform in the ratio of raw materials-reagent (1:10) in a Soxhlet apparatus. The resulting extract was concentrated under mild conditions to a thick concentrate, which was analyzed on a gas chromatograph with a mass selective detector. The analyzes were carried out on a gas chromatograph with a mass spectrometric detector 6890N / 5973C (Agilent, USA). The peaks detected in the chromatograms were identified using the NIST’11 and Wiley 10 mass spectral libraries. The article first studied and conducted a comparative analysis of lipophilic substances contained in the fruits of the plant Rosa canina L, harvested in the Almaty region. The extracts mainly include hydrocarbons, derivatives of higher carboxylic acids, higher alcohols and some other substances, in particular butyl tetradecyl ether of hydrochloric acid, [4- (2-methylpentanoylsulfamoyl) phenyl] amide of 2-methylpentanoic acid in hexane extract and, butyl undecyl ether sulfuric acid, eicosyl trifluoroacetate and ethoxycarbonyl-3-methyl-4-azafluorenone, 2-fluorenilim in the chloroform extract. Further work will be continued in order to determine the potential biological activity of the obtained extracts. Keywords: Rosa canina L., СКФ-экстракция, никотин, GC-MS, химический состав.

In this article identifies new sources of obtaining biological substances from plants of the genus Atraphaxis (A. spinosa) prepared in the Almaty region. According to well - known methods, the analysis of indicators and standards of raw material quality-humidity, total ash, sulphate ash, insoluble ash in 10% hydrochloric acid-was developed and carried out. Micro- and macroelements determined by atomic absorption spectroscopy. Analysis of the elemental composition shows that iron predominates from microelements, and sodium, potassium and calcium from macronutrients. Conditions for obtaining a biologically active complex from the aboveground part of Atraphaxis spinosa developed for the first time. The optimal conditions for obtaining the complex are extractant – 50% ethanol, the ratio of extractant and raw materials – 1:8, double extraction time – 48 hours, temperature – 22 - 26 °C. The lipophilic composition identified by chromatography-mass spectroscopy. Since lipophilic fractions of plant samples include such classes of compounds as fatty acids; mono-; di-; triglycerides, phospholipids, sterols, Sterol esters, glycolipids, fat-soluble vitamins, they can considered not only as nutritional products, but also as possible pharmacological agents. The content of lipophilic components – 26 organic compounds-was determined. It found that A. spinosa contains a large amount of di - (2-ethylhexyl) phthalate (54.66%) and β-sitosterol (13.11%). A complex study of plant resources as medicinal raw materials provides for the chemical study of biologically active substances and biological screening of extracts and individual compounds obtained from plants. In most cases, the extract showed a wide range of antibacterial activity against the used strains of microorganisms.

The paper discusses a comprehensive analysis of contemporary approaches to the structural modification of phenolic acids from the benzoic (C₆–C₁) and cinnamic (C₆–C₃) series, which have a wide range of biological activity. The main directions of chemical transformation of phenolic compounds that aim to improve their pharmaco-logical potential, stability, and bioavailability are discussed. The given study also makes an accent on the relationship between the structural features and biological effects of various phenolic acid derivatives, including antioxidant, anti-inflammatory, antimicrobial, and antitumor activity. The purpose of this investigation is to systematize current data on strategies for the structural modification of phenolic acids, to identify key areas of their chemical transformation, and to determine the most promising methods for creating biologically active compounds with improved pharmacological properties. The study was aimed at systematizing the accumulated knowledge and identifying promising areas for further research in the field of structural design of phenolic acid derivatives. Based on the analysis of experimental data and literature sources, key trends in the development of new medicinal compounds using the example of natural phenolic acids were characterized, and prospects for their use in medicine and pharmaceutical chemistry were shown.

This study systematically investigated lipophilic and polar metabolites of Atraphaxis virgata (Polygonaceae) and assessed its immunomodulatory activity in vivo. Supercritical CO₂ extraction of the aerial parts yielded a lipophilic fraction analyzed by means of gas chromatography–mass spectrometry (GC–MS), which identified 42 compounds, including fatty acid esters, sterols, hydrocarbons, and terpenoids. The residual plant meal was subjected to ultrasound-assisted extraction with 70% aqueous ethanol at 30–35 °C, using a solid-to-solvent ratio of 1:8 for 120 min. This polar extract was evaluated for amino acids, proteins, and carbohydrates, while solvent–solvent partitioning with chloroform, ethyl acetate, and water enabled isolation of phenolic- and flavonoid-enriched fractions. Six phenolic constituents, including four flavonol glycosides and two phenolic acids, were structurally confirmed. The extracts were rich in unsaturated fatty acids and water-soluble antioxidants, supporting their nutritional and pharmacological relevance. In vivo evaluation using a cyclophosphamide-induced myelosuppression model in Wistar rats demonstrated stimulation of erythropoiesis and leukopoiesis, confirming immunomodulatory potential. Collectively, this work provides the first comprehensive chemical and biological characterization of A. virgata and establishes a foundation for mechanistic studies and pharmacological validation.

This study investigates the proximate composition, mineral content, and fatty acid profile of Canavalia ensiformis L. beans extracted with various solvents. Proximate analysis revealed a high protein content (23.16%), moderate fiber levels (7.13%), and a total fat content of 2.35%. Elemental analysis identified potassium (18,696.87 mg/kg) as the most abundant macroelement, followed by calcium (1,886.81 mg/kg) and magnesium (1,500.98 mg/kg), whereas iron (50.43 mg/kg) and zinc (33.28 mg/kg) were the predominant microelements. Gas chromatography analysis of fatty acids showed that monounsaturated fatty acids (MUFA) dominated the lipid profile, particularly in the hexane (58.67%), ethanol (59.57%), and raw bean (61.11%) extracts. Oleic acid (C18:1n9c) was the most abundant MUFA, reaching 51.39% in the hexane extract. Saturated fatty acids (SFA) were highest in the hexane extract (21.35%), primarily due to palmitic acid (C16:0, 18.77%). Polyunsaturated fatty acids (PUFA) were most prevalent in the hexane extract (16.23%) and bean sample (16.08%), with linoleic acid (C18:2n6c) being the major component. The aqueous extract contained notable amounts of γ-linolenic acid (C18:3n6, 1.68%). These findings highlight the nutritional potential of C. ensiformis beans and suggest that solvent selection significantly influences lipid composition. The high MUFA and PUFA content, particularly in hexane and ethanol extracts, underscores their possible applications in food and pharmaceutical formulations. Key words: Canavalia ensiformis L., GC analysis, proximate analysis, micro- and macroelements, fatty acids.