Density Functional Theory in Computer-Aided Drug Design: A Systematic Review of Applications, Methodological Trends, and Integration with Molecular Modeling Workflows

Density Functional Theory in Computer-Aided Drug Design: A Systematic Review of Applications, Methodological Trends, and Integration with Molecular Modeling Workflows | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Systematic Review Density Functional Theory in Computer-Aided Drug Design: A Systematic Review of Applications, Methodological Trends, and Integration with Molecular Modeling Workflows Frans Augusthinus Asmuruf, Supeno Supeno, Yuliana Ruth Yabansabra, and 2 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-9151448/v1 This work is licensed under a CC BY 4.0 License Status: Under Review Version 1 posted 4 You are reading this latest preprint version Abstract Density Functional Theory (DFT) is becoming an increasingly valuable tool in computer-aided molecular design because it provides quantum-level insight that can strengthen practical drug discovery workflows. Beyond its traditional role in electronic structure analysis, DFT is now widely used to support reactivity assessment, binding interpretation, and molecular descriptor generation within integrated in silico strategies. This systematic review examines how DFT has been applied in recent drug discovery research, with particular attention to methodological trends, therapeutic targets, and its integration with other computational approaches. Following PRISMA 2020 guidelines, studies published between January 2021 and February 2025 were identified through searches of Google Scholar, PubMed/MEDLINE, Web of Science, Scopus, and IEEE Xplore. Of 815 records screened, 81 studies met the inclusion criteria. B3LYP and 6-31G(d,p) remained the most commonly used functional and basis set, appearing in 67.9% and 71.6% of studies, respectively. DFT was most often combined with molecular docking (76.5%), ADMET prediction (55.6%), and molecular dynamics (34.6%), highlighting its growing role in multi-method drug design workflows. The main application areas were infectious diseases, cancer, and metabolic disorders. Overall, the evidence suggests that DFT is no longer used primarily as a stand-alone quantum chemical method, but rather as an integrated component of modern molecular design pipelines that supports mechanistic understanding and rational compound optimization. Future advances will likely come from closer integration with machine learning, better treatment of conformational flexibility, and more realistic solvent-aware simulations. Density Functional Theory computer-aided molecular design drug discovery molecular dynamics ADMET PRISMA Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Figure 7 Figure 8 Figure 9 1. Introduction 1.1 Background and Rationale The growing integration of computational methods into pharmaceutical research has fundamentally changed how new drug candidates are identified, evaluated, and optimized. Among these methods, Density Functional Theory (DFT) has become especially valuable because it provides molecular-level insight into electronic structure, reactivity, intermolecular interactions, and structure–property relationships that are directly relevant to rational drug design. Since the foundational work of Hohenberg and Kohn, followed by the practical formulation introduced by Kohn and Sham, DFT has progressed from a largely theoretical quantum mechanical framework to an important component of modern computer-aided molecular design [ 1 ]–[ 3 ]. The practical relevance of DFT has become even more apparent as drug discovery has shifted toward integrated in silico workflows. Rather than being used in isolation, DFT is increasingly combined with molecular docking, ADMET prediction, molecular dynamics, and other computational approaches to improve mechanistic interpretation and support more informed decision-making during lead identification and optimization. In this context, DFT contributes a level of electronic and energetic detail that complements the speed and screening power of broader molecular modeling techniques [ 1 ]. The importance of computational chemistry was particularly evident during the COVID-19 pandemic, when rapid, cost-effective approaches were urgently needed to evaluate potential therapeutic candidates. DFT contributed to these efforts by helping researchers assess molecular stability, electronic reactivity, and binding-related properties of antiviral compounds, using docking and other computational tools. Studies on SARS-CoV-2-related compounds demonstrated how DFT could be incorporated into practical drug discovery workflows that balanced computational efficiency with deeper chemical insight [ 4 ]–[ 7 ]. This period helped reinforce the value of DFT not only as a theoretical method but also as a flexible and actionable tool in applied pharmaceutical research. At the same time, the rapid expansion of DFT-based studies has created a need for a more systematic evaluation of how the method is currently being used in the literature. Although many reports have demonstrated the utility of DFT in medicinal chemistry, fewer studies have critically examined broader methodological patterns, including the choice of functionals and basis sets, the extent of integration with complementary in silico methods, the therapeutic areas most frequently targeted, and the recurring limitations that affect reproducibility and predictive value. Without such synthesis, it remains difficult to determine whether the growing use of DFT reflects methodological maturity, routine convention, or genuine strategic integration into computer-aided drug design workflows. This systematic review was therefore designed to examine recent applications of DFT in drug discovery and molecular modeling from a methodological and strategic perspective. By analyzing studies published between 2021 and early 2025, this review aims to identify dominant computational practices, evaluate integration patterns with other in silico techniques, map key therapeutic application areas, and highlight current limitations and future opportunities. In doing so, it seeks to clarify the evolving role of DFT within contemporary computer-aided molecular design and to provide a structured evidence base for researchers developing more robust and predictive computational drug discovery pipelines. Accordingly, this systematic review was undertaken to clarify how DFT is currently being used within contemporary drug discovery and molecular modeling research. Specifically, the review aims to identify dominant methodological choices, including the selection of functionals, basis sets, and software platforms; examine how DFT is integrated with complementary in silico approaches such as molecular docking, ADMET prediction, molecular dynamics, and machine learning; map the principal therapeutic application areas; and evaluate recurring limitations related to validation, reproducibility, solvation treatment, and conformational sampling. By synthesizing recent evidence from 2021 to early 2025, this review seeks to provide a structured and up-to-date perspective on the evolving role of DFT in computer-aided molecular design and to highlight priorities for building more robust, predictive, and mechanistically informed computational drug discovery workflows. 2. Methods 2.1 Protocol and Registration This systematic review was conducted according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) 2020 guidelines [ 8 ]. 2.2 Search Strategy Comprehensive literature searches were conducted across five major databases: Google Scholar, PubMed/MEDLINE, Web of Science, Scopus, and IEEE Xplore. The search strategy employed Boolean operators to combine concept clusters covering DFT methodologies, in silico approaches, and pharmaceutical applications. The final search was conducted on February 26, 2025. 2.3 Eligibility Criteria Inclusion Criteria : Original research articles or reviews published in Q1–Q4 indexed journals Publication date between January 1, 2021, and February 26, 2025 Studies employing DFT calculations for pharmaceutical or biological applications Full-text availability in English Exclusion Criteria : Conference abstracts, editorials, or non-peer-reviewed publications Studies not indexed in recognized journal ranking systems Publications lacking an explicit DFT methodology description 2.4 Study Selection and Data Extraction Study selection was performed in two stages using the predefined inclusion and exclusion criteria. Title and abstract screening was first conducted using Rayyan QCRI [ 9 ], followed by full-text assessment of potentially eligible studies. Any uncertainties regarding study eligibility were resolved through discussion among the authors. Data from the included studies were then collected to summarize publication characteristics, therapeutic application, computational methods, DFT settings, software used, integration with other in silico techniques, and validation approach. 2.5 Quality Assessment Study quality was evaluated using an author-adapted methodological quality rubric for computational studies, broadly informed by established risk-of-bias and reporting frameworks such as PROBAST, assessing methodological transparency (0–3 points), validation approach (0–3 points), reproducibility (0–2 points), and statistical rigor (0–2 points), yielding total scores ranging from 0 to 10 [ 10 ]. 3. Results 3.1 Study Selection and Characteristics The PRISMA flow diagram (Fig. 1 ) illustrates the systematic selection process. Of 815 unique records identified after duplicate removal, 160 full-text articles were assessed for eligibility, and 81 studies met the inclusion criteria. Figure 2 presents three analyses of studies. Figure 2 (a) shows the publication trend for 2021–2025, indicating sustained research activity, with a peak in 2023 (27.2% of total publications), coinciding with the post-pandemic consolidation of computational methodologies. Figure 2 (b) displays the citation distribution, with 39.5% of studies receiving 0–20 citations and 7.4% achieving high impact (> 100 citations), indicating significant variability in research influence. Figure 2 (c) shows that focus areas highlight drug design applications (55.6% of studies), followed by molecular docking integration (47.5%) and antiviral/COVID-19 research (30.9%). This distribution reflects both the methodological maturity of DFT in pharmaceutical applications and the responsiveness of computational methods to global health priorities. 3.2 Bibliometric Analysis Figure 3 provides a detailed analysis of citation metrics. Figure 3 (a) shows the citation distribution histogram, revealing a right-skewed pattern typical of emerging research fields, with most studies clustering in the 11–50 citation range. The mean citation count of 40.0 (SD = 42.3) substantially exceeds typical values for specialized computational chemistry reviews, suggesting strong field impact. Figure 3 (b) illustrates citation trends by publication year, demonstrating an expected decline in mean citations for more recent publications (2024–2025), while median values remain more stable. This pattern reflects the time-dependent nature of citation accumulation, with earlier publications having a greater opportunity to be cited. The top 5 most cited articles (Fig. 3 (c)) include Bitew et al.’s [ 11 ] study on antidiabetic flavonoids (217 citations), followed by Noureddine et al.’s [ 6 ] COVID-19 research (145 citations). These high-impact studies share common characteristics: integration of multiple computational methods, relevance to major disease areas, and publication in accessible journals. Figure 4 examines journal distribution patterns. Figure 4 (a) shows that 46.9% of studies were published in Q1 journals, with an additional 34.6% in Q2 journals, indicating high overall publication quality. The top publication venues (Fig. 4 (b)) include the Journal of Molecular Structure (7.4% of studies), Scientific Reports (4.9%), and International Journal of Molecular Sciences (3.7%), reflecting the interdisciplinary nature of DFT-based pharmaceutical research. The journal impact factor distribution (Fig. 4 (c)) reveals that 39.5% of studies appeared in journals with IFs of 3–5, while 29.6% had Ifs > 5. Figure 4 (d) demonstrates relatively consistent publication quality over the years, with Q1 publications maintaining a steady share throughout the review period. 3.3 Methodological Characteristics Figure 5 analyzes computational software and methodology choices. Figure 5 (a) shows Gaussian 09/16 dominance (59.3% of studies), reflecting its established reliability and extensive institutional licensing. However, the significant adoption of ORCA (19.8%) and GAMESS (12.3%) indicates growing awareness of accessibility, as these free alternatives offer comparable functionality across many applications [ 12 ]. Software usage trends (Fig. 5 (b)) reveal a decline in Gaussian dominance in recent years, potentially indicating methodological diversification or cost-driven adoption of free alternatives. The DFT functionals distribution (Fig. 5 (c)) confirms B3LYP’s predominance (67.9%). However, the 22.2% adoption of M06-2X and 14.8% for ωB97X-D suggests increasing sophistication in functional selection for systems requiring dispersion correction or charge-transfer description [ 13 ]. Basis set selection (Fig. 5 (d)) shows 71.6% utilization of the 6-31G(d,p) basis, with 27.2% using the larger 6-311G(d,p) basis. Notably, only 18.5% incorporated diffuse functions (6–31 + G(d,p)), potentially compromising accuracy for anionic species or hydrogen-bonded systems common in drug molecules [ 14 ]. 3.4 Application Domains Figure 6 explores therapeutic applications and biological targets. Figure 6 (a) shows infectious diseases as the dominant therapeutic area (34.6%), driven primarily by COVID-19 research (22.2% of all studies). Cancer research (27.2%) and metabolic disorders (22.2%) represent other major application domains, reflecting global disease priorities and research funding patterns. The temporal analysis (Fig. 6 (b)) reveals 2023 as the peak year for infectious disease publications, coinciding with pandemic response efforts, whereas cancer research shows a more consistent yearly distribution. Biological target classes (Fig. 6 (c)) show enzyme predominance (55.6%), followed by viral proteins (22.2%) and receptors (14.8%), which aligns with the distribution of therapeutic areas. Specific diseases studied (Fig. 6 (d)) highlight COVID-19 (18 studies), diabetes (12), and breast cancer (8) as primary foci, with Alzheimer’s disease and malaria also receiving significant attention. This distribution reflects both global health priorities and the suitability of DFT methods for studying the molecular mechanisms underlying these conditions. 3.5 Integration with Complementary Methods Figure 7 examines the use of molecular docking and molecular dynamics software. Figure 7 (a) shows AutoDock Vina dominance (48.4% of docking studies), attributable to its open-source nature, favorable speed-accuracy balance, and extensive community support [ 15 ]. AutoDock 4.2 maintains significant usage (24.2%), while commercial packages (Glide, GOLD, MOE) collectively represent 26.4%. The distribution of docking scoring functions (Fig. 7 (b)) reveals that Vina Score predominates, with traditional AutoDock scoring and Glide Score also widely used. This diversity suggests ongoing evaluation of scoring function performance across different target classes. Molecular dynamics software (Fig. 7 (c)) shows GROMACS leadership (50.0% of MD studies), reflecting its free availability, extensive feature set, and strong performance on modern GPU architectures [ 16 ]. AMBER (25.0%) and NAMD (14.3%) maintain significant user bases, particularly for specialized applications requiring specific force fields or enhanced sampling methods. Figure 7 (d) shows that simulation times vary considerably, with 50–100 ns being most common (35.3% of MD studies), though substantial proportions employ shorter (< 50 ns, 23.5%) or longer (100–200 ns, 23.5%) simulations. This variability suggests a lack of standardized protocols, with simulation length often dictated by computational resources rather than convergence requirements. 3.6 ADMET Prediction and Machine Learning Figure 8 analyzes ADMET prediction tools and their integration with machine learning. Figure 8 (a) shows SwissADME as the dominant platform (47.1% of ADMET studies), attributable to its user-friendly web interface, comprehensive descriptor set, and free accessibility [ 17 ]. pkCSM (17.6%) and ADMET Predictor (13.7%) represent other major tools, with commercial alternatives maintaining significant market presence. Figure 8 (b) displays predicted ADMET parameters, revealing that toxicity assessment is the most common (94.1% of ADMET studies), followed by drug-likeness evaluation (100%) and metabolism prediction (82.4%). This distribution reflects regulatory requirements and the high cost of late-stage toxicity failures in drug development. Machine learning integration (Fig. 8 (c)) remains limited (18.5% of all studies), with Random Forest as the predominant method (40.0% of ML studies), followed by Support Vector Machines (20.0%) and Neural Networks (13.3%). The emerging adoption of Deep Learning (13.3%) suggests growing interest in end-to-end learning approaches, though dataset size limitations may constrain applicability [ 18 ]. ML application areas (Fig. 8 (d)) show activity prediction as the primary use case (33.3%), followed by property prediction (26.7%) and virtual screening (20.0%). The relatively low overall adoption of ML suggests the field remains in the early stages of integrating these approaches with DFT calculations. 3.7 Quality Assessment Results Figure 9 presents a comprehensive quality assessment analysis. Figure 9 (a) shows the quality score distribution, with 56.8% of studies achieving high quality (≥ 7/10) and 43.2% moderate quality (4–6/10). Notably, no studies fell into the low-quality category (< 4/10), indicating overall methodological rigor in the field. Quality by criterion (Fig. 9 (b)) reveals methodological transparency as the strongest aspect (mean 2.3/3, 77% compliance), while validation approaches (mean 1.8/3, 60%) and statistical rigor (mean 1.2/2, 60%) show room for improvement. This pattern suggests studies generally report computational parameters adequately but often lack experimental validation or appropriate statistical analyses. Temporal quality trends (Fig. 9 (c)) show a consistent high-quality representation across years, with 2023 showing the highest proportion of high-quality studies (63.6%). This stability suggests established methodological standards rather than a decline in quality with increasing publication volume. The quality summary statistics (Fig. 9 (d)) highlight common limitations: insufficient experimental validation (59.3% of studies), lack of statistical analysis (54.3%), and incomplete reproducibility information (43.2%). These findings align with broader concerns about reproducibility in computational chemistry and suggest specific areas for methodological improvement [ 19 ]. 4. Discussion 4.1 Principal Findings This systematic review of 81 DFT-based in silico studies reveals a mature, diverse, and rapidly evolving field with significant impact on pharmaceutical research. The analysis identifies several key trends as follows. Methodological Standardization The predominance of B3LYP/6-31G(d,p) combinations (67.9%/71.6%) provides consistency across studies, though opportunities exist to optimize the functional level based on specific application requirements. The significant adoption of dispersion-corrected functionals (37.0% of studies) indicates growing awareness of B3LYP’s limitations for non-covalent interactions [ 20 ]. Integration Imperative The high frequency of multi-method approaches (76.5% with docking, 34.6% with MD) demonstrates recognition that DFT alone cannot capture the full complexity of biological systems. The prominence of docking-integrated studies is consistent with the established role of molecular docking in structure-based drug discovery [ 21 ]. Pandemic Responsiveness The substantial COVID-19 research component (22.2% of all studies) illustrates the agility of computational methods in addressing emergent health threats. However, limited experimental validation in this domain (33.3% of COVID-19 studies) raises concerns about predictive reliability [ 22 ]. Quality Standards The absence of low-quality studies and 56.8% high-quality representation indicates established methodological standards. However, validation gaps (59.3% without experimental data) and statistical limitations (54.3% lacking appropriate analyses) suggest areas for improvement. 4.2 Methodological Considerations The analysis of computational approaches reveals several important considerations, as discussed below. Basis Set Selection The reliance on 6-31G(d,p) (71.6%) may limit accuracy for highly polar or anionic systems. The underutilization of diffuse functions (18.5%), therefore, represents a methodological gap, particularly for ligands containing carboxylate or phosphate groups [ 23 ] Solvation Effects Only 64.0% of studies explicitly accounted for solvent effects using Polarizable Continuum Models. Given that biological systems are aqueous environments, this omission may limit the physiological relevance of gas-phase calculations [ 24 ]. Validation Practices The 40.7% of studies including experimental validation, while substantial, suggest room for improvement. In purely computational studies, benchmarking against higher-level electronic-structure methods can provide an additional layer of confidence in methodological accuracy [ 25 ]. 4.3 Therapeutic Area Analysis The therapeutic distribution reflects global disease priorities and research funding patterns. The prominence of infectious disease research (34.6%), particularly COVID-19 studies, demonstrates the responsiveness of computational methods to emerging health crises. However, the limited experimental validation in this domain suggests the need for post-computational experimental follow-up. Cancer drug discovery applications (27.2%) show consistent integration of DFT with HDAC inhibitor design and metal-based anticancer agents, demonstrating the method’s versatility across diverse chemical spaces. The correlation between DFT-derived electrophilicity and cytotoxicity provides a mechanistic rationale for compound optimization [ 26 ]. 4.4 Integration Strategies The combination of DFT electronic descriptors with molecular docking scores improves predictive accuracy for binding affinity, as electronic complementarity augments geometric fit. Studies employing this integrated approach showed higher citation impact, suggesting greater perceived value in the research community. The incorporation of ADMET predictions (55.6% of studies) represents a shift toward early-stage pharmacokinetic assessment, potentially reducing late-stage drug development failures. The dominance of SwissADME (47.1%) reflects its accessibility and comprehensive descriptor set, though validation against experimental pharmacokinetic data remains essential [ 27 ]. The limited but growing adoption of machine learning (18.5%) reflects a broader shift toward data-driven molecular design. In this context, DFT-derived descriptors can serve as informative features for predictive models, although interpretability remains an important consideration [ 28 ]. 4.5 Limitations and Future Directions Several systematic limitations were identified: Validation Gaps Only 40.7% of studies included experimental validation, limiting assessment of predictive accuracy. Future studies should prioritize experimental follow-up or benchmark against established datasets. Statistical Rigor 54.3% of studies lacked appropriate statistical analyses of descriptor-activity correlations, potentially leading to spurious associations. Robust statistical frameworks should be standard practice. Reproducibility 43.2% of studies did not provide sufficient methodological detail to enable straightforward reproduction. Complete reporting of computational parameters, model setup, and analysis workflow should therefore be standard practice [ 29 ]. Conformational Sampling 51.9% of studies employed a single optimized conformation, which may not adequately represent the conformational flexibility relevant to ligand behavior in complex biological environments. Broader conformational exploration should therefore be considered where feasible [ 30 ]. Emerging trends likely to shape future research include double-hybrid functionals for improved accuracy, greater use of explicit-solvent simulations and ab initio molecular dynamics to capture solvent-specific effects, quantum computing applications for exponential speedup, and automated workflows that integrate DFT with robotic synthesis [ 31 ]–[ 34 ]. 5. Conclusions This PRISMA-compliant systematic review of 81 DFT-based in silico studies (2021–2025) establishes the methodology as a cornerstone of modern pharmaceutical research. Key conclusions include: 1. Methodological Maturity Standardization around B3LYP/6-31G(d,p) provides consistency, though opportunities exist to optimize functionals for specific applications, particularly for dispersion-dominated systems. 2. Integration Imperative The most prominent studies combine DFT with molecular docking (76.5%), ADMET prediction (55.6%), and increasingly, molecular dynamics (34.6%) and machine learning (18.5%). 3. Therapeutic Versatility Applications span infectious diseases (34.6%), cancer (27.2%), metabolic disorders (22.2%), and emerging areas, demonstrating broad applicability. 4. Quality Standards While 56.8% of studies achieved high quality, validation gaps and statistical limitations suggest areas for methodological improvement. 5. Future Trajectory The convergence of DFT with machine learning, quantum computing, and automated experimentation positions the methodology for continued growth and impact. The field stands at an inflection point: established enough to have standardized protocols, yet dynamic enough to incorporate emerging technologies. Realizing the full potential of DFT in drug discovery requires addressing validation gaps, enhancing reproducibility, and fostering interdisciplinary collaboration between computational chemists, medicinal chemists, and data scientists. Declarations Ethics Approval and Consent to Participate Not applicable (systematic review of published literature). Consent for Publication Not applicable. Competing Interests The authors declare no competing interests. Funding This research was supported by the Graduate School of Cenderawasih University under Grant No. 1671/UN20.2.1/PG/2025. Author Contribution All authors contributed to the study conception and design. Material preparation, data collection, and analysis were performed by Frans Augusthinus Asmuruf, Supeno, Ilham Salim, Yuliana Ruth Yabansabra, and Eva Susanty Simaremare. Frans Augusthinus Asmuruf wrote the first draft of the manuscript, and all authors commented on previous versions of the manuscript. All authors read and approved the final manuscript. Acknowledgement We would like to thank the Graduate School of the University of Cenderawasih for the funding. Data Availability All data generated or analyzed during this study are included in this published article and its supplementary information files. References de Freitas Rodrigues SB, de Araújo RSA, de Mendonça TRD, Mendonça-Júnior FJB, Zhan P, da Silva-Júnior EF (2023) Quantum Chemistry in Drug Design: Density Function Theory (DFT) and Other Quantum Mechanics (QM)-Related Approaches. in Applied Computer-Aided Drug Design: Models and Methods. Bentham Science, Singapore, pp 187–220. https://doi.org/10.2174/9789815179934123010010 . Hohenberg P, Kohn W (1964) Inhomogeneous electron gas. Phys Rev 136(3B):B864–B871. https://doi.org/10.1103/PhysRev.136.B864 Kohn W, Sham LJ (1965) Self-consistent equations including exchange and correlation effects. Phys Rev 140(4A):A1133–A1138. https://doi.org/10.1103/PhysRev.140.A1133 Ghosh R, Chakraborty A, Biswas A, Chowdhuri S (2021) Evaluation of green tea polyphenols as novel coronavirus 3CL protease inhibitors. J Biomol Struct Dyn 39(18):6972–6985. https://doi.org/10.1080/07391102.2020.1779818 Kashyap J, Datta D (2022) Drug repurposing for SARS-CoV-2: a high-throughput molecular docking, molecular dynamics, machine learning, and DFT study. J Mater Sci 57(23):10780–10802. https://doi.org/10.1007/s10853-022-07195-8 Noureddine O, Issaoui S, Ghalla H (2021) Quantum chemical studies on molecular structure, AIM, ELF, RDG and antiviral activities of hybrid hydroxychloroquine in the treatment of COVID-19. J King Saud Univ Sci 33(1):101334. https://doi.org/10.1016/j.jksus.2020.101334 Erdogan T (2021) DFT, molecular docking and molecular dynamics simulation studies on some newly introduced natural products for their potential use against SARS-CoV-2. J Mol Struct 1242:130733. https://doi.org/10.1016/j.molstruc.2021.130733 Page MJ, McKenzie JE, Bossuyt PM et al (2021) The PRISMA 2020 statement: an updated guideline for reporting systematic reviews. BMJ 372:n71. https://doi.org/10.1136/bmj.n71 Ouzzani M, Hammady H, Fedorowicz Z, Elmagarmid A (2016) Rayyan—a web and mobile app for systematic reviews. Syst Rev 5(1):210. https://doi.org/10.1186/s13643-016-0384-4 Wolff RF, Moons KGM, Riley RD et al (2019) PROBAST: A tool to assess the risk of bias and applicability of prediction model studies. Ann Intern Med 170(1):51–58. https://doi.org/10.7326/M18-1376 Bitew M, Desalegn T, Demissie TB, Belayneh A, Endale M, Eswaramoorthy R (2021) Pharmacokinetics and drug-likeness of antidiabetic flavonoids: Molecular docking and DFT study. PLoS ONE 16(12):e0260853. https://doi.org/10.1371/journal.pone.0260853 Neese F (2018) Software update: The ORCA program system, version 4.0. WIREs Comput Mol Sci 8(1):e1327. https://doi.org/10.1002/wcms.1327 Zhao Y, Truhlar DG (2008) The M06 suite of density functionals for main group thermochemistry, thermochemical kinetics, noncovalent interactions, excited states, and transition elements. Theor Chem Acc 120:215–241. https://doi.org/10.1007/s00214-007-0310-x Clark T, Chandrasekhar J, Spitznagel GW, Schleyer PvR (1983) Efficient diffuse function-augmented basis sets for anion calculations. III. The 3–21 + G basis set for first-row elements, Li–F. J Comput Chem 4(3):294–301. https://doi.org/10.1002/jcc.540040303 Trott O, Olson AJ (2010) AutoDock Vina: improving the speed and accuracy of docking with a new scoring function, efficient optimization, and multithreading. J Comput Chem 31(2):455–461. https://doi.org/10.1002/jcc.21334 Abraham MJ, Murtola T, Schulz R et al (2015) GROMACS: High performance molecular simulations through multi-level parallelism from laptops to supercomputers. SoftwareX 1–2:19–25. https://doi.org/10.1016/j.softx.2015.06.001 Daina A, Michielin O, Zoete V (2017) SwissADME: a free web tool to evaluate pharmacokinetics, drug-likeness and medicinal chemistry friendliness of small molecules. Sci Rep 7:42717. https://doi.org/10.1038/srep42717 Butler KT, Davies DW, Cartwright H, Isayev O, Walsh A (2018) Machine learning for molecular and materials science. Nature 559(7715):547–555. https://doi.org/10.1038/s41586-018-0337-2 Wilkinson MD, Dumontier M, Aalbersberg IJ et al (2016) The FAIR Guiding Principles for scientific data management and stewardship. Sci Data 3:160018. https://doi.org/10.1038/sdata.2016.18 Grimme S (2006) Semiempirical GGA-type density functional constructed with a long-range dispersion correction. J Comput Chem 27(15):1787–1799. https://doi.org/10.1002/jcc.20495 Meng X-Y, Zhang H-X, Mezei M, Cui M (2011) Molecular docking: a powerful approach for structure-based drug discovery. Curr Comput -Aided Drug Des 7(2):146–157. https://doi.org/10.2174/157340911795677602 Mishra SS, Rai SK, Raghuwanshi HS et al (2021) Computational investigation of potential inhibitors of novel coronavirus 2019 through structure-based virtual screening, molecular dynamics and density functional theory studies. J Biomol Struct Dyn 39(12):4449–4461. https://doi.org/10.1080/07391102.2020.1791957 Hariharan PC, Pople JA (1973) The influence of polarization functions on molecular orbital hydrogenation energies. Theor Chim Acta 28(3):213–222. https://doi.org/10.1007/BF00533485 Tomasi J, Mennucci B, Cammi R (2005) Quantum mechanical continuum solvation models. Chem Rev 105(8):2999–3093. https://doi.org/10.1021/cr9904009 Goerigk L, Grimme S (2011) Efficient and accurate double-hybrid-meta-GGA density functionals—evaluation with the extended GMTKN30 database for general main group thermochemistry, kinetics, and noncovalent interactions. J Chem Theory Comput 7(2):291–309. https://doi.org/10.1021/ct100466k Pal R, Patra SG, Chattaraj PK (2022) Quantitative structure–toxicity relationship in bioactive molecules from a conceptual DFT perspective. Pharmaceuticals 15(11):1383. https://doi.org/10.3390/ph15111383 Hughes JP, Rees S, Kalindjian SB, Philpott KL (2011) Principles of early drug discovery. Br J Pharmacol 162(6):1239–1249. https://doi.org/10.1111/j.1476-5381.2010.01127.x Goldsmith BR, Esterhuizen J, Liu J-X, Bartel CJ, Sutton C (2018) Machine learning for heterogeneous catalyst design and discovery. AIChE J 64(7):2311–2323. https://doi.org/10.1002/aic.16198 Leach AR (2001) Molecular Modelling: Principles and Applications , 2nd ed. Harlow: Prentice Hall, ISBN: 978-0582382107 Lane TJ, Shukla D, Beauchamp KA, Pande VS (2013) To milliseconds and beyond: challenges in the simulation of protein folding. Curr Opin Struct Biol 23(1):58–65. https://doi.org/10.1016/j.sbi.2012.11.002 Grimme S (2006) Semiempirical hybrid density functional with perturbative second-order correlation. J Chem Phys 124(3):034108. https://doi.org/10.1063/1.2148954 Cao Y, Romero J, Olson JP et al (2019) Quantum chemistry in the age of quantum computing. Chem Rev 119(19):1085610915. https://doi.org/10.1021/acs.chemrev.8b00803 Caramelli D, Granda S, Cambié L et al (2021) Discovering new chemistry with an autonomous robotic platform. ACS Cent Sci 7(11):1821–1830. https://doi.org/10.1021/acscentsci.1c00435 Zhang H, Juraskova V, Duarte F (2024) Modelling chemical processes in explicit solvents with machine learning potentials. Nat Commun 15(1):6114. https://doi.org/10.1038/s41467-024-50418-6 Additional Declarations No competing interests reported. Supplementary Files SupplementaryMaterial1CompleteStudies.xlsx SupplementaryMaterial2DataExtractionForms.xlsx SupplementaryMaterial3PRISMAChecklist.xlsx SupplementaryMaterial4QualityAssessment.xlsx SupplementaryMaterial5RawDataAnalysis.xlsx Cite Share Download PDF Status: Under Review Version 1 posted Reviewers invited by journal 30 Mar, 2026 Editor assigned by journal 30 Mar, 2026 Submission checks completed at journal 18 Mar, 2026 First submitted to journal 17 Mar, 2026 You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-9151448","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Systematic Review","associatedPublications":[],"authors":[{"id":608254009,"identity":"795d51ed-921a-421b-bddf-a959f3a35f46","order_by":0,"name":"Frans Augusthinus Asmuruf","email":"data:image/png;base64,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","orcid":"","institution":"Cenderawasih University","correspondingAuthor":true,"prefix":"","firstName":"Frans","middleName":"Augusthinus","lastName":"Asmuruf","suffix":""},{"id":608254010,"identity":"2e55269f-1ac0-4a4c-bd74-c042e5afea71","order_by":1,"name":"Supeno Supeno","email":"","orcid":"","institution":"Cenderawasih University","correspondingAuthor":false,"prefix":"","firstName":"Supeno","middleName":"","lastName":"Supeno","suffix":""},{"id":608254011,"identity":"4af7d0da-8cdd-4747-96ce-f4c6c5034690","order_by":2,"name":"Yuliana Ruth Yabansabra","email":"","orcid":"","institution":"Cenderawasih University","correspondingAuthor":false,"prefix":"","firstName":"Yuliana","middleName":"Ruth","lastName":"Yabansabra","suffix":""},{"id":608254013,"identity":"9a826e38-4293-45db-920f-62f6628cdd34","order_by":3,"name":"Ilham Salim","email":"","orcid":"","institution":"Cenderawasih University","correspondingAuthor":false,"prefix":"","firstName":"Ilham","middleName":"","lastName":"Salim","suffix":""},{"id":608254014,"identity":"0d9bcb7e-3375-46a9-b13d-ce9257b89abb","order_by":4,"name":"Eva Susanty Simaremare","email":"","orcid":"","institution":"Cenderawasih University","correspondingAuthor":false,"prefix":"","firstName":"Eva","middleName":"Susanty","lastName":"Simaremare","suffix":""}],"badges":[],"createdAt":"2026-03-17 17:24:11","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-9151448/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-9151448/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":105562896,"identity":"09deeb2c-33f6-4a35-b3ed-73d503c9bc78","added_by":"auto","created_at":"2026-03-27 12:45:10","extension":"jpeg","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":524779,"visible":true,"origin":"","legend":"\u003cp\u003ePRISMA 2020 flow diagram illustrating the systematic selection process for DFT in silico studies (2021–2025).\u003c/p\u003e","description":"","filename":"floatimage1.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-9151448/v1/72391dd74666999c56306f8e.jpeg"},{"id":105727778,"identity":"cd065c54-1344-4bd5-97c7-f8ddebbe325b","added_by":"auto","created_at":"2026-03-30 11:03:34","extension":"jpeg","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":190433,"visible":true,"origin":"","legend":"\u003cp\u003ePRISMA 2020 flow diagram illustrating the systematic selection process for DFT in silico studies (2021–2025).\u003c/p\u003e","description":"","filename":"floatimage2.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-9151448/v1/2b33f49df86eeed003584dad.jpeg"},{"id":105562988,"identity":"570cc2c2-c1a8-4e9c-bdfe-4543e5bd19b9","added_by":"auto","created_at":"2026-03-27 12:45:32","extension":"jpeg","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":385695,"visible":true,"origin":"","legend":"\u003cp\u003e(a) Citation distribution, (b) Citation trends by year, (c) Top five most cited articles\u003c/p\u003e","description":"","filename":"floatimage3.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-9151448/v1/c1008a18de07cce0f0b031f2.jpeg"},{"id":105563206,"identity":"f3a859b6-dd09-4b2d-b9aa-8447f7cd5433","added_by":"auto","created_at":"2026-03-27 12:46:20","extension":"jpeg","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":43027,"visible":true,"origin":"","legend":"\u003cp\u003e(a) Journal quartile distribution, (b) Top publication venues, (c) Journal impact factor distribution, (d) Publication trend by journal quartile.\u003c/p\u003e","description":"","filename":"groupimage1.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-9151448/v1/0ad90d8b737317640bea73f3.jpeg"},{"id":105563273,"identity":"81785fcb-6791-4e74-b12a-1b4b6cf3e2a9","added_by":"auto","created_at":"2026-03-27 12:46:36","extension":"jpeg","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":50488,"visible":true,"origin":"","legend":"\u003cp\u003e(a) Computational software usage, (b) Software usage trends, (c) DFT functionals distribution, (d) Basis sets usage.\u003c/p\u003e","description":"","filename":"groupimage2.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-9151448/v1/82f26fc7a54fdf1a5896819f.jpeg"},{"id":105562919,"identity":"68d23e2f-ccfe-410f-bcd7-5f7063989ed1","added_by":"auto","created_at":"2026-03-27 12:45:16","extension":"jpeg","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":52582,"visible":true,"origin":"","legend":"\u003cp\u003e(a) Therapeutic areas distribution, (b) Top 3 therapeutic areas by year, (c) Biological target classes, (d) Specific diseases studied.\u003c/p\u003e","description":"","filename":"groupimage3.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-9151448/v1/5623888908fb325e70800ae9.jpeg"},{"id":105084150,"identity":"021b7205-1433-4c27-a1b5-88481d0a7fcd","added_by":"auto","created_at":"2026-03-20 19:03:27","extension":"jpeg","order_by":7,"title":"Figure 7","display":"","copyAsset":false,"role":"figure","size":43257,"visible":true,"origin":"","legend":"\u003cp\u003e(a) Molecular docking software, (b) Docking scoring function, (c) Molecular dynamics software, (d) MD simulation time.\u003c/p\u003e","description":"","filename":"groupimage4.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-9151448/v1/5947876001c62cc0b7bed54d.jpeg"},{"id":105084151,"identity":"d67401f2-b40c-4f85-b50c-53a18ed45050","added_by":"auto","created_at":"2026-03-20 19:03:27","extension":"jpeg","order_by":8,"title":"Figure 8","display":"","copyAsset":false,"role":"figure","size":46399,"visible":true,"origin":"","legend":"\u003cp\u003e(a) ADMET prediction tools, (b) ADMET parameters predicted, (c) Machine learning method, (d) ML application areas.\u003c/p\u003e","description":"","filename":"groupimage5.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-9151448/v1/f9ed3940a18c243c069140a9.jpeg"},{"id":105563214,"identity":"ce2bffbd-870a-495e-b74a-5f3af4dd4ef0","added_by":"auto","created_at":"2026-03-27 12:46:21","extension":"jpeg","order_by":9,"title":"Figure 9","display":"","copyAsset":false,"role":"figure","size":60601,"visible":true,"origin":"","legend":"\u003cp\u003e(a) Quality score distribution, (b) Quality assessment by criteria, (c) Quality distribution by year, (d) Quality assessment summary.\u003c/p\u003e","description":"","filename":"groupimage6.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-9151448/v1/87b170edeef087cbd1ff7e5c.jpeg"},{"id":105729798,"identity":"5f307f0c-b386-430b-a421-18fe67a11c02","added_by":"auto","created_at":"2026-03-30 11:20:07","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":2231407,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-9151448/v1/b79f400b-372e-4803-b385-9e341e6680c6.pdf"},{"id":105562965,"identity":"efdcd224-9ead-426f-877a-8828db3e5309","added_by":"auto","created_at":"2026-03-27 12:45:26","extension":"xlsx","order_by":0,"title":"","display":"","copyAsset":false,"role":"supplement","size":27272,"visible":true,"origin":"","legend":"","description":"","filename":"SupplementaryMaterial1CompleteStudies.xlsx","url":"https://assets-eu.researchsquare.com/files/rs-9151448/v1/074892624b84351b0cb25ed4.xlsx"},{"id":105563230,"identity":"6cb4e6ad-dc50-4d0d-898b-acb6165b378e","added_by":"auto","created_at":"2026-03-27 12:46:26","extension":"xlsx","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":42313,"visible":true,"origin":"","legend":"","description":"","filename":"SupplementaryMaterial2DataExtractionForms.xlsx","url":"https://assets-eu.researchsquare.com/files/rs-9151448/v1/9465f1f303e683b966b53750.xlsx"},{"id":105084145,"identity":"33eb54c0-7984-4ee9-b797-928d69f4354e","added_by":"auto","created_at":"2026-03-20 19:03:27","extension":"xlsx","order_by":2,"title":"","display":"","copyAsset":false,"role":"supplement","size":9088,"visible":true,"origin":"","legend":"","description":"","filename":"SupplementaryMaterial3PRISMAChecklist.xlsx","url":"https://assets-eu.researchsquare.com/files/rs-9151448/v1/fda52770d08a75c2c16bbd5f.xlsx"},{"id":105563269,"identity":"6b8d775d-36bd-4a38-97dc-df051757106b","added_by":"auto","created_at":"2026-03-27 12:46:36","extension":"xlsx","order_by":3,"title":"","display":"","copyAsset":false,"role":"supplement","size":26719,"visible":true,"origin":"","legend":"","description":"","filename":"SupplementaryMaterial4QualityAssessment.xlsx","url":"https://assets-eu.researchsquare.com/files/rs-9151448/v1/445ab9ee3b1c7ae896faf3c7.xlsx"},{"id":105084154,"identity":"86afec81-c71b-401c-8b0a-4c72d28f4cbc","added_by":"auto","created_at":"2026-03-20 19:03:27","extension":"xlsx","order_by":4,"title":"","display":"","copyAsset":false,"role":"supplement","size":42811,"visible":true,"origin":"","legend":"","description":"","filename":"SupplementaryMaterial5RawDataAnalysis.xlsx","url":"https://assets-eu.researchsquare.com/files/rs-9151448/v1/08b6ccce05ba8082779547e8.xlsx"}],"financialInterests":"No competing interests reported.","formattedTitle":"Density Functional Theory in Computer-Aided Drug Design: A Systematic Review of Applications, Methodological Trends, and Integration with Molecular Modeling Workflows","fulltext":[{"header":"1. Introduction","content":"\u003cdiv id=\"Sec2\" class=\"Section2\"\u003e \u003ch2\u003e1.1 Background and Rationale\u003c/h2\u003e \u003cp\u003eThe growing integration of computational methods into pharmaceutical research has fundamentally changed how new drug candidates are identified, evaluated, and optimized. Among these methods, Density Functional Theory (DFT) has become especially valuable because it provides molecular-level insight into electronic structure, reactivity, intermolecular interactions, and structure\u0026ndash;property relationships that are directly relevant to rational drug design. Since the foundational work of Hohenberg and Kohn, followed by the practical formulation introduced by Kohn and Sham, DFT has progressed from a largely theoretical quantum mechanical framework to an important component of modern computer-aided molecular design [\u003cspan additionalcitationids=\"CR2\" citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e]\u0026ndash;[\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eThe practical relevance of DFT has become even more apparent as drug discovery has shifted toward integrated in silico workflows. Rather than being used in isolation, DFT is increasingly combined with molecular docking, ADMET prediction, molecular dynamics, and other computational approaches to improve mechanistic interpretation and support more informed decision-making during lead identification and optimization. In this context, DFT contributes a level of electronic and energetic detail that complements the speed and screening power of broader molecular modeling techniques [\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eThe importance of computational chemistry was particularly evident during the COVID-19 pandemic, when rapid, cost-effective approaches were urgently needed to evaluate potential therapeutic candidates. DFT contributed to these efforts by helping researchers assess molecular stability, electronic reactivity, and binding-related properties of antiviral compounds, using docking and other computational tools. Studies on SARS-CoV-2-related compounds demonstrated how DFT could be incorporated into practical drug discovery workflows that balanced computational efficiency with deeper chemical insight [\u003cspan additionalcitationids=\"CR5 CR6\" citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e]\u0026ndash;[\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e]. This period helped reinforce the value of DFT not only as a theoretical method but also as a flexible and actionable tool in applied pharmaceutical research.\u003c/p\u003e \u003cp\u003eAt the same time, the rapid expansion of DFT-based studies has created a need for a more systematic evaluation of how the method is currently being used in the literature. Although many reports have demonstrated the utility of DFT in medicinal chemistry, fewer studies have critically examined broader methodological patterns, including the choice of functionals and basis sets, the extent of integration with complementary in silico methods, the therapeutic areas most frequently targeted, and the recurring limitations that affect reproducibility and predictive value. Without such synthesis, it remains difficult to determine whether the growing use of DFT reflects methodological maturity, routine convention, or genuine strategic integration into computer-aided drug design workflows.\u003c/p\u003e \u003cp\u003eThis systematic review was therefore designed to examine recent applications of DFT in drug discovery and molecular modeling from a methodological and strategic perspective. By analyzing studies published between 2021 and early 2025, this review aims to identify dominant computational practices, evaluate integration patterns with other in silico techniques, map key therapeutic application areas, and highlight current limitations and future opportunities. In doing so, it seeks to clarify the evolving role of DFT within contemporary computer-aided molecular design and to provide a structured evidence base for researchers developing more robust and predictive computational drug discovery pipelines.\u003c/p\u003e \u003cp\u003eAccordingly, this systematic review was undertaken to clarify how DFT is currently being used within contemporary drug discovery and molecular modeling research. Specifically, the review aims to identify dominant methodological choices, including the selection of functionals, basis sets, and software platforms; examine how DFT is integrated with complementary in silico approaches such as molecular docking, ADMET prediction, molecular dynamics, and machine learning; map the principal therapeutic application areas; and evaluate recurring limitations related to validation, reproducibility, solvation treatment, and conformational sampling. By synthesizing recent evidence from 2021 to early 2025, this review seeks to provide a structured and up-to-date perspective on the evolving role of DFT in computer-aided molecular design and to highlight priorities for building more robust, predictive, and mechanistically informed computational drug discovery workflows.\u003c/p\u003e \u003c/div\u003e"},{"header":"2. Methods","content":"\u003cdiv id=\"Sec4\" class=\"Section2\"\u003e \u003ch2\u003e2.1 Protocol and Registration\u003c/h2\u003e \u003cp\u003eThis systematic review was conducted according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) 2020 guidelines [\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e].\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec5\" class=\"Section2\"\u003e \u003ch2\u003e2.2 Search Strategy\u003c/h2\u003e \u003cp\u003eComprehensive literature searches were conducted across five major databases: Google Scholar, PubMed/MEDLINE, Web of Science, Scopus, and IEEE Xplore. The search strategy employed Boolean operators to combine concept clusters covering DFT methodologies, in silico approaches, and pharmaceutical applications. The final search was conducted on February 26, 2025.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec6\" class=\"Section2\"\u003e \u003ch2\u003e2.3 Eligibility Criteria\u003c/h2\u003e \u003cp\u003e \u003cem\u003eInclusion Criteria\u003c/em\u003e:\u003c/p\u003e \u003cp\u003e \u003cul\u003e \u003cli\u003e \u003cp\u003eOriginal research articles or reviews published in Q1\u0026ndash;Q4 indexed journals\u003c/p\u003e \u003c/li\u003e \u003cli\u003e \u003cp\u003ePublication date between January 1, 2021, and February 26, 2025\u003c/p\u003e \u003c/li\u003e \u003cli\u003e \u003cp\u003eStudies employing DFT calculations for pharmaceutical or biological applications\u003c/p\u003e \u003c/li\u003e \u003cli\u003e \u003cp\u003eFull-text availability in English\u003c/p\u003e \u003c/li\u003e \u003c/ul\u003e \u003c/p\u003e \u003cp\u003e \u003cem\u003eExclusion Criteria\u003c/em\u003e:\u003c/p\u003e \u003cp\u003e \u003cul\u003e \u003cli\u003e \u003cp\u003eConference abstracts, editorials, or non-peer-reviewed publications\u003c/p\u003e \u003c/li\u003e \u003cli\u003e \u003cp\u003eStudies not indexed in recognized journal ranking systems\u003c/p\u003e \u003c/li\u003e \u003cli\u003e \u003cp\u003ePublications lacking an explicit DFT methodology description\u003c/p\u003e \u003c/li\u003e \u003c/ul\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec7\" class=\"Section2\"\u003e \u003ch2\u003e2.4 Study Selection and Data Extraction\u003c/h2\u003e \u003cp\u003eStudy selection was performed in two stages using the predefined inclusion and exclusion criteria. Title and abstract screening was first conducted using Rayyan QCRI [\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e], followed by full-text assessment of potentially eligible studies. Any uncertainties regarding study eligibility were resolved through discussion among the authors. Data from the included studies were then collected to summarize publication characteristics, therapeutic application, computational methods, DFT settings, software used, integration with other in silico techniques, and validation approach.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec8\" class=\"Section2\"\u003e \u003ch2\u003e2.5 Quality Assessment\u003c/h2\u003e \u003cp\u003eStudy quality was evaluated using an author-adapted methodological quality rubric for computational studies, broadly informed by established risk-of-bias and reporting frameworks such as PROBAST, assessing methodological transparency (0\u0026ndash;3 points), validation approach (0\u0026ndash;3 points), reproducibility (0\u0026ndash;2 points), and statistical rigor (0\u0026ndash;2 points), yielding total scores ranging from 0 to 10 [\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e].\u003c/p\u003e \u003c/div\u003e"},{"header":"3. Results","content":"\u003cdiv id=\"Sec10\" class=\"Section2\"\u003e \u003ch2\u003e3.1 Study Selection and Characteristics\u003c/h2\u003e \u003cp\u003eThe PRISMA flow diagram (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e) illustrates the systematic selection process. Of 815 unique records identified after duplicate removal, 160 full-text articles were assessed for eligibility, and 81 studies met the inclusion criteria.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eFigure \u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e presents three analyses of studies. Figure\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e (a) shows the publication trend for 2021\u0026ndash;2025, indicating sustained research activity, with a peak in 2023 (27.2% of total publications), coinciding with the post-pandemic consolidation of computational methodologies. Figure\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e (b) displays the citation distribution, with 39.5% of studies receiving 0\u0026ndash;20 citations and 7.4% achieving high impact (\u0026gt;\u0026thinsp;100 citations), indicating significant variability in research influence. Figure\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e (c) shows that focus areas highlight drug design applications (55.6% of studies), followed by molecular docking integration (47.5%) and antiviral/COVID-19 research (30.9%). This distribution reflects both the methodological maturity of DFT in pharmaceutical applications and the responsiveness of computational methods to global health priorities.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec11\" class=\"Section2\"\u003e \u003ch2\u003e3.2 Bibliometric Analysis\u003c/h2\u003e \u003cp\u003eFigure \u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003e provides a detailed analysis of citation metrics. Figure\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003e (a) shows the citation distribution histogram, revealing a right-skewed pattern typical of emerging research fields, with most studies clustering in the 11\u0026ndash;50 citation range. The mean citation count of 40.0 (SD\u0026thinsp;=\u0026thinsp;42.3) substantially exceeds typical values for specialized computational chemistry reviews, suggesting strong field impact. Figure\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003e (b) illustrates citation trends by publication year, demonstrating an expected decline in mean citations for more recent publications (2024\u0026ndash;2025), while median values remain more stable. This pattern reflects the time-dependent nature of citation accumulation, with earlier publications having a greater opportunity to be cited. The top 5 most cited articles (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003e (c)) include Bitew et al.\u0026rsquo;s [\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e] study on antidiabetic flavonoids (217 citations), followed by Noureddine et al.\u0026rsquo;s [\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e] COVID-19 research (145 citations). These high-impact studies share common characteristics: integration of multiple computational methods, relevance to major disease areas, and publication in accessible journals.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eFigure \u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003e examines journal distribution patterns. Figure\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003e (a) shows that 46.9% of studies were published in Q1 journals, with an additional 34.6% in Q2 journals, indicating high overall publication quality.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eThe top publication venues (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003e (b)) include the Journal of Molecular Structure (7.4% of studies), Scientific Reports (4.9%), and International Journal of Molecular Sciences (3.7%), reflecting the interdisciplinary nature of DFT-based pharmaceutical research. The journal impact factor distribution (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003e (c)) reveals that 39.5% of studies appeared in journals with IFs of 3\u0026ndash;5, while 29.6% had Ifs\u0026thinsp;\u0026gt;\u0026thinsp;5. Figure\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003e (d) demonstrates relatively consistent publication quality over the years, with Q1 publications maintaining a steady share throughout the review period.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec12\" class=\"Section2\"\u003e \u003ch2\u003e3.3 Methodological Characteristics\u003c/h2\u003e \u003cp\u003eFigure \u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003e analyzes computational software and methodology choices. Figure\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003e (a) shows Gaussian 09/16 dominance (59.3% of studies), reflecting its established reliability and extensive institutional licensing. However, the significant adoption of ORCA (19.8%) and GAMESS (12.3%) indicates growing awareness of accessibility, as these free alternatives offer comparable functionality across many applications [\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e].\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eSoftware usage trends (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003e (b)) reveal a decline in Gaussian dominance in recent years, potentially indicating methodological diversification or cost-driven adoption of free alternatives. The DFT functionals distribution (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003e (c)) confirms B3LYP\u0026rsquo;s predominance (67.9%). However, the 22.2% adoption of M06-2X and 14.8% for ωB97X-D suggests increasing sophistication in functional selection for systems requiring dispersion correction or charge-transfer description [\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eBasis set selection (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003e (d)) shows 71.6% utilization of the 6-31G(d,p) basis, with 27.2% using the larger 6-311G(d,p) basis. Notably, only 18.5% incorporated diffuse functions (6\u0026ndash;31\u0026thinsp;+\u0026thinsp;G(d,p)), potentially compromising accuracy for anionic species or hydrogen-bonded systems common in drug molecules [\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e].\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec13\" class=\"Section2\"\u003e \u003ch2\u003e3.4 Application Domains\u003c/h2\u003e \u003cp\u003eFigure \u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003e explores therapeutic applications and biological targets. Figure\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003e (a) shows infectious diseases as the dominant therapeutic area (34.6%), driven primarily by COVID-19 research (22.2% of all studies). Cancer research (27.2%) and metabolic disorders (22.2%) represent other major application domains, reflecting global disease priorities and research funding patterns.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eThe temporal analysis (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003e (b)) reveals 2023 as the peak year for infectious disease publications, coinciding with pandemic response efforts, whereas cancer research shows a more consistent yearly distribution. Biological target classes (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003e (c)) show enzyme predominance (55.6%), followed by viral proteins (22.2%) and receptors (14.8%), which aligns with the distribution of therapeutic areas. Specific diseases studied (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003e (d)) highlight COVID-19 (18 studies), diabetes (12), and breast cancer (8) as primary foci, with Alzheimer\u0026rsquo;s disease and malaria also receiving significant attention. This distribution reflects both global health priorities and the suitability of DFT methods for studying the molecular mechanisms underlying these conditions.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec14\" class=\"Section2\"\u003e \u003ch2\u003e3.5 Integration with Complementary Methods\u003c/h2\u003e \u003cp\u003eFigure \u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003e examines the use of molecular docking and molecular dynamics software. Figure\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003e (a) shows AutoDock Vina dominance (48.4% of docking studies), attributable to its open-source nature, favorable speed-accuracy balance, and extensive community support [\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e]. AutoDock 4.2 maintains significant usage (24.2%), while commercial packages (Glide, GOLD, MOE) collectively represent 26.4%.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eThe distribution of docking scoring functions (Fig.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003e (b)) reveals that Vina Score predominates, with traditional AutoDock scoring and Glide Score also widely used. This diversity suggests ongoing evaluation of scoring function performance across different target classes. Molecular dynamics software (Fig.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003e (c)) shows GROMACS leadership (50.0% of MD studies), reflecting its free availability, extensive feature set, and strong performance on modern GPU architectures [\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e]. AMBER (25.0%) and NAMD (14.3%) maintain significant user bases, particularly for specialized applications requiring specific force fields or enhanced sampling methods. Figure\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003e (d) shows that simulation times vary considerably, with 50\u0026ndash;100 ns being most common (35.3% of MD studies), though substantial proportions employ shorter (\u0026lt;\u0026thinsp;50 ns, 23.5%) or longer (100\u0026ndash;200 ns, 23.5%) simulations. This variability suggests a lack of standardized protocols, with simulation length often dictated by computational resources rather than convergence requirements.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec15\" class=\"Section2\"\u003e \u003ch2\u003e3.6 ADMET Prediction and Machine Learning\u003c/h2\u003e \u003cp\u003eFigure \u003cspan refid=\"Fig8\" class=\"InternalRef\"\u003e8\u003c/span\u003e analyzes ADMET prediction tools and their integration with machine learning. Figure\u0026nbsp;\u003cspan refid=\"Fig8\" class=\"InternalRef\"\u003e8\u003c/span\u003e (a) shows SwissADME as the dominant platform (47.1% of ADMET studies), attributable to its user-friendly web interface, comprehensive descriptor set, and free accessibility [\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e]. pkCSM (17.6%) and ADMET Predictor (13.7%) represent other major tools, with commercial alternatives maintaining significant market presence.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eFigure \u003cspan refid=\"Fig8\" class=\"InternalRef\"\u003e8\u003c/span\u003e (b) displays predicted ADMET parameters, revealing that toxicity assessment is the most common (94.1% of ADMET studies), followed by drug-likeness evaluation (100%) and metabolism prediction (82.4%). This distribution reflects regulatory requirements and the high cost of late-stage toxicity failures in drug development. Machine learning integration (Fig.\u0026nbsp;\u003cspan refid=\"Fig8\" class=\"InternalRef\"\u003e8\u003c/span\u003e (c)) remains limited (18.5% of all studies), with Random Forest as the predominant method (40.0% of ML studies), followed by Support Vector Machines (20.0%) and Neural Networks (13.3%). The emerging adoption of Deep Learning (13.3%) suggests growing interest in end-to-end learning approaches, though dataset size limitations may constrain applicability [\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e]. ML application areas (Fig.\u0026nbsp;\u003cspan refid=\"Fig8\" class=\"InternalRef\"\u003e8\u003c/span\u003e (d)) show activity prediction as the primary use case (33.3%), followed by property prediction (26.7%) and virtual screening (20.0%). The relatively low overall adoption of ML suggests the field remains in the early stages of integrating these approaches with DFT calculations.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec16\" class=\"Section2\"\u003e \u003ch2\u003e3.7 Quality Assessment Results\u003c/h2\u003e \u003cp\u003eFigure \u003cspan refid=\"Fig9\" class=\"InternalRef\"\u003e9\u003c/span\u003e presents a comprehensive quality assessment analysis. Figure\u0026nbsp;\u003cspan refid=\"Fig9\" class=\"InternalRef\"\u003e9\u003c/span\u003e (a) shows the quality score distribution, with 56.8% of studies achieving high quality (\u0026ge;\u0026thinsp;7/10) and 43.2% moderate quality (4\u0026ndash;6/10). Notably, no studies fell into the low-quality category (\u0026lt;\u0026thinsp;4/10), indicating overall methodological rigor in the field. Quality by criterion (Fig.\u0026nbsp;\u003cspan refid=\"Fig9\" class=\"InternalRef\"\u003e9\u003c/span\u003e (b)) reveals methodological transparency as the strongest aspect (mean 2.3/3, 77% compliance), while validation approaches (mean 1.8/3, 60%) and statistical rigor (mean 1.2/2, 60%) show room for improvement. This pattern suggests studies generally report computational parameters adequately but often lack experimental validation or appropriate statistical analyses. Temporal quality trends (Fig.\u0026nbsp;\u003cspan refid=\"Fig9\" class=\"InternalRef\"\u003e9\u003c/span\u003e (c)) show a consistent high-quality representation across years, with 2023 showing the highest proportion of high-quality studies (63.6%). This stability suggests established methodological standards rather than a decline in quality with increasing publication volume. The quality summary statistics (Fig.\u0026nbsp;\u003cspan refid=\"Fig9\" class=\"InternalRef\"\u003e9\u003c/span\u003e (d)) highlight common limitations: insufficient experimental validation (59.3% of studies), lack of statistical analysis (54.3%), and incomplete reproducibility information (43.2%). These findings align with broader concerns about reproducibility in computational chemistry and suggest specific areas for methodological improvement [\u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e].\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e"},{"header":"4. Discussion","content":"\u003cdiv id=\"Sec18\" class=\"Section2\"\u003e \u003ch2\u003e4.1 Principal Findings\u003c/h2\u003e \u003cp\u003eThis systematic review of 81 DFT-based in silico studies reveals a mature, diverse, and rapidly evolving field with significant impact on pharmaceutical research. The analysis identifies several key trends as follows.\u003c/p\u003e \u003cp\u003e \u003cem\u003eMethodological Standardization\u003c/em\u003e \u003c/p\u003e \u003cp\u003eThe predominance of B3LYP/6-31G(d,p) combinations (67.9%/71.6%) provides consistency across studies, though opportunities exist to optimize the functional level based on specific application requirements. The significant adoption of dispersion-corrected functionals (37.0% of studies) indicates growing awareness of B3LYP\u0026rsquo;s limitations for non-covalent interactions [\u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e].\u003c/p\u003e \u003cp\u003e \u003cem\u003eIntegration Imperative\u003c/em\u003e \u003c/p\u003e \u003cp\u003eThe high frequency of multi-method approaches (76.5% with docking, 34.6% with MD) demonstrates recognition that DFT alone cannot capture the full complexity of biological systems. The prominence of docking-integrated studies is consistent with the established role of molecular docking in structure-based drug discovery [\u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e].\u003c/p\u003e \u003cp\u003e \u003cem\u003ePandemic Responsiveness\u003c/em\u003e \u003c/p\u003e \u003cp\u003eThe substantial COVID-19 research component (22.2% of all studies) illustrates the agility of computational methods in addressing emergent health threats. However, limited experimental validation in this domain (33.3% of COVID-19 studies) raises concerns about predictive reliability [\u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e].\u003c/p\u003e \u003cp\u003e \u003cem\u003eQuality Standards\u003c/em\u003e \u003c/p\u003e \u003cp\u003eThe absence of low-quality studies and 56.8% high-quality representation indicates established methodological standards. However, validation gaps (59.3% without experimental data) and statistical limitations (54.3% lacking appropriate analyses) suggest areas for improvement.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec19\" class=\"Section2\"\u003e \u003ch2\u003e4.2 Methodological Considerations\u003c/h2\u003e \u003cp\u003eThe analysis of computational approaches reveals several important considerations, as discussed below.\u003c/p\u003e \u003cp\u003e \u003cem\u003eBasis Set Selection\u003c/em\u003e \u003c/p\u003e \u003cp\u003eThe reliance on 6-31G(d,p) (71.6%) may limit accuracy for highly polar or anionic systems. The underutilization of diffuse functions (18.5%), therefore, represents a methodological gap, particularly for ligands containing carboxylate or phosphate groups [\u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e]\u003c/p\u003e \u003cp\u003e \u003cem\u003eSolvation Effects\u003c/em\u003e \u003c/p\u003e \u003cp\u003eOnly 64.0% of studies explicitly accounted for solvent effects using Polarizable Continuum Models. Given that biological systems are aqueous environments, this omission may limit the physiological relevance of gas-phase calculations [\u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e].\u003c/p\u003e \u003cp\u003e \u003cem\u003eValidation Practices\u003c/em\u003e \u003c/p\u003e \u003cp\u003eThe 40.7% of studies including experimental validation, while substantial, suggest room for improvement. In purely computational studies, benchmarking against higher-level electronic-structure methods can provide an additional layer of confidence in methodological accuracy [\u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e].\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec20\" class=\"Section2\"\u003e \u003ch2\u003e4.3 Therapeutic Area Analysis\u003c/h2\u003e \u003cp\u003eThe therapeutic distribution reflects global disease priorities and research funding patterns. The prominence of infectious disease research (34.6%), particularly COVID-19 studies, demonstrates the responsiveness of computational methods to emerging health crises. However, the limited experimental validation in this domain suggests the need for post-computational experimental follow-up. Cancer drug discovery applications (27.2%) show consistent integration of DFT with HDAC inhibitor design and metal-based anticancer agents, demonstrating the method\u0026rsquo;s versatility across diverse chemical spaces. The correlation between DFT-derived electrophilicity and cytotoxicity provides a mechanistic rationale for compound optimization [\u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e26\u003c/span\u003e].\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec21\" class=\"Section2\"\u003e \u003ch2\u003e4.4 Integration Strategies\u003c/h2\u003e \u003cp\u003eThe combination of DFT electronic descriptors with molecular docking scores improves predictive accuracy for binding affinity, as electronic complementarity augments geometric fit. Studies employing this integrated approach showed higher citation impact, suggesting greater perceived value in the research community. The incorporation of ADMET predictions (55.6% of studies) represents a shift toward early-stage pharmacokinetic assessment, potentially reducing late-stage drug development failures. The dominance of SwissADME (47.1%) reflects its accessibility and comprehensive descriptor set, though validation against experimental pharmacokinetic data remains essential [\u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e27\u003c/span\u003e]. The limited but growing adoption of machine learning (18.5%) reflects a broader shift toward data-driven molecular design. In this context, DFT-derived descriptors can serve as informative features for predictive models, although interpretability remains an important consideration [\u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e28\u003c/span\u003e].\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec22\" class=\"Section2\"\u003e \u003ch2\u003e4.5 Limitations and Future Directions\u003c/h2\u003e \u003cp\u003eSeveral systematic limitations were identified:\u003c/p\u003e \u003cp\u003e \u003cem\u003eValidation Gaps\u003c/em\u003e \u003c/p\u003e \u003cp\u003eOnly 40.7% of studies included experimental validation, limiting assessment of predictive accuracy. Future studies should prioritize experimental follow-up or benchmark against established datasets.\u003c/p\u003e \u003cp\u003e \u003cem\u003eStatistical Rigor\u003c/em\u003e \u003c/p\u003e \u003cp\u003e54.3% of studies lacked appropriate statistical analyses of descriptor-activity correlations, potentially leading to spurious associations. Robust statistical frameworks should be standard practice.\u003c/p\u003e \u003cp\u003e \u003cem\u003eReproducibility\u003c/em\u003e \u003c/p\u003e \u003cp\u003e43.2% of studies did not provide sufficient methodological detail to enable straightforward reproduction. Complete reporting of computational parameters, model setup, and analysis workflow should therefore be standard practice [\u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e29\u003c/span\u003e].\u003c/p\u003e \u003cp\u003e \u003cem\u003eConformational Sampling\u003c/em\u003e \u003c/p\u003e \u003cp\u003e51.9% of studies employed a single optimized conformation, which may not adequately represent the conformational flexibility relevant to ligand behavior in complex biological environments. Broader conformational exploration should therefore be considered where feasible [\u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e30\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eEmerging trends likely to shape future research include double-hybrid functionals for improved accuracy, greater use of explicit-solvent simulations and ab initio molecular dynamics to capture solvent-specific effects, quantum computing applications for exponential speedup, and automated workflows that integrate DFT with robotic synthesis [\u003cspan additionalcitationids=\"CR32 CR33\" citationid=\"CR31\" class=\"CitationRef\"\u003e31\u003c/span\u003e]\u0026ndash;[\u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e34\u003c/span\u003e].\u003c/p\u003e \u003c/div\u003e"},{"header":"5. Conclusions","content":"\u003cp\u003eThis PRISMA-compliant systematic review of 81 DFT-based in silico studies (2021\u0026ndash;2025) establishes the methodology as a cornerstone of modern pharmaceutical research. Key conclusions include:\u003c/p\u003e\n\u003ch3\u003e1. Methodological Maturity\u003c/h3\u003e\n\u003cp\u003e \u003cdiv class=\"BlockQuote\"\u003e \u003cp\u003eStandardization around B3LYP/6-31G(d,p) provides consistency, though opportunities exist to optimize functionals for specific applications, particularly for dispersion-dominated systems.\u003c/p\u003e \u003c/div\u003e \u003c/p\u003e\n\u003ch3\u003e2. Integration Imperative\u003c/h3\u003e\n\u003cp\u003e \u003cdiv class=\"BlockQuote\"\u003e \u003cp\u003eThe most prominent studies combine DFT with molecular docking (76.5%), ADMET prediction (55.6%), and increasingly, molecular dynamics (34.6%) and machine learning (18.5%).\u003c/p\u003e \u003c/div\u003e \u003c/p\u003e\n\u003ch3\u003e3. Therapeutic Versatility\u003c/h3\u003e\n\u003cp\u003e \u003cdiv class=\"BlockQuote\"\u003e \u003cp\u003eApplications span infectious diseases (34.6%), cancer (27.2%), metabolic disorders (22.2%), and emerging areas, demonstrating broad applicability.\u003c/p\u003e \u003c/div\u003e \u003c/p\u003e\n\u003ch3\u003e4. Quality Standards\u003c/h3\u003e\n\u003cp\u003e \u003cdiv class=\"BlockQuote\"\u003e \u003cp\u003eWhile 56.8% of studies achieved high quality, validation gaps and statistical limitations suggest areas for methodological improvement.\u003c/p\u003e \u003c/div\u003e \u003c/p\u003e\n\u003ch3\u003e5. Future Trajectory\u003c/h3\u003e\n\u003cp\u003e \u003cdiv class=\"BlockQuote\"\u003e \u003cp\u003eThe convergence of DFT with machine learning, quantum computing, and automated experimentation positions the methodology for continued growth and impact.\u003c/p\u003e \u003c/div\u003e \u003c/p\u003e \u003cp\u003eThe field stands at an inflection point: established enough to have standardized protocols, yet dynamic enough to incorporate emerging technologies. Realizing the full potential of DFT in drug discovery requires addressing validation gaps, enhancing reproducibility, and fostering interdisciplinary collaboration between computational chemists, medicinal chemists, and data scientists.\u003c/p\u003e"},{"header":"Declarations","content":" \u003cp\u003e \u003cstrong\u003eEthics Approval and Consent to Participate\u003c/strong\u003e \u003cp\u003eNot applicable (systematic review of published literature).\u003c/p\u003e \u003c/p\u003e \u003cp\u003e \u003cstrong\u003eConsent for Publication\u003c/strong\u003e \u003cp\u003eNot applicable.\u003c/p\u003e \u003c/p\u003e\u003cp\u003e \u003ch2\u003eCompeting Interests\u003c/h2\u003e \u003cp\u003eThe authors declare no competing interests.\u003c/p\u003e \u003c/p\u003e\u003ch2\u003eFunding\u003c/h2\u003e \u003cp\u003eThis research was supported by the Graduate School of Cenderawasih University under Grant No. 1671/UN20.2.1/PG/2025.\u003c/p\u003e\u003ch2\u003eAuthor Contribution\u003c/h2\u003e\u003cp\u003eAll authors contributed to the study conception and design. Material preparation, data collection, and analysis were performed by Frans Augusthinus Asmuruf, Supeno, Ilham Salim, Yuliana Ruth Yabansabra, and Eva Susanty Simaremare. Frans Augusthinus Asmuruf wrote the first draft of the manuscript, and all authors commented on previous versions of the manuscript. All authors read and approved the final manuscript.\u003c/p\u003e\u003ch2\u003eAcknowledgement\u003c/h2\u003e\u003cp\u003eWe would like to thank the Graduate School of the University of Cenderawasih for the funding.\u003c/p\u003e\u003ch2\u003eData Availability\u003c/h2\u003e\u003cp\u003eAll data generated or analyzed during this study are included in this published article and its supplementary information files.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\u003cli\u003e\u003cspan\u003ede Freitas Rodrigues SB, de Ara\u0026uacute;jo RSA, de Mendon\u0026ccedil;a TRD, Mendon\u0026ccedil;a-J\u0026uacute;nior FJB, Zhan P, da Silva-J\u0026uacute;nior EF (2023) Quantum Chemistry in Drug Design: Density Function Theory (DFT) and Other Quantum Mechanics (QM)-Related Approaches. in Applied Computer-Aided Drug Design: Models and Methods. Bentham Science, Singapore, pp 187\u0026ndash;220. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.2174/9789815179934123010010\u003c/span\u003e\u003cspan address=\"10.2174/9789815179934123010010\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHohenberg P, Kohn W (1964) Inhomogeneous electron gas. Phys Rev 136(3B):B864\u0026ndash;B871. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1103/PhysRev.136.B864\u003c/span\u003e\u003cspan address=\"10.1103/PhysRev.136.B864\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKohn W, Sham LJ (1965) Self-consistent equations including exchange and correlation effects. Phys Rev 140(4A):A1133\u0026ndash;A1138. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1103/PhysRev.140.A1133\u003c/span\u003e\u003cspan address=\"10.1103/PhysRev.140.A1133\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGhosh R, Chakraborty A, Biswas A, Chowdhuri S (2021) Evaluation of green tea polyphenols as novel coronavirus 3CL protease inhibitors. J Biomol Struct Dyn 39(18):6972\u0026ndash;6985. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1080/07391102.2020.1779818\u003c/span\u003e\u003cspan address=\"10.1080/07391102.2020.1779818\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKashyap J, Datta D (2022) Drug repurposing for SARS-CoV-2: a high-throughput molecular docking, molecular dynamics, machine learning, and DFT study. J Mater Sci 57(23):10780\u0026ndash;10802. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1007/s10853-022-07195-8\u003c/span\u003e\u003cspan address=\"10.1007/s10853-022-07195-8\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eNoureddine O, Issaoui S, Ghalla H (2021) Quantum chemical studies on molecular structure, AIM, ELF, RDG and antiviral activities of hybrid hydroxychloroquine in the treatment of COVID-19. J King Saud Univ Sci 33(1):101334. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.jksus.2020.101334\u003c/span\u003e\u003cspan address=\"10.1016/j.jksus.2020.101334\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eErdogan T (2021) DFT, molecular docking and molecular dynamics simulation studies on some newly introduced natural products for their potential use against SARS-CoV-2. J Mol Struct 1242:130733. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.molstruc.2021.130733\u003c/span\u003e\u003cspan address=\"10.1016/j.molstruc.2021.130733\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePage MJ, McKenzie JE, Bossuyt PM et al (2021) The PRISMA 2020 statement: an updated guideline for reporting systematic reviews. BMJ 372:n71. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1136/bmj.n71\u003c/span\u003e\u003cspan address=\"10.1136/bmj.n71\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eOuzzani M, Hammady H, Fedorowicz Z, Elmagarmid A (2016) Rayyan\u0026mdash;a web and mobile app for systematic reviews. Syst Rev 5(1):210. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1186/s13643-016-0384-4\u003c/span\u003e\u003cspan address=\"10.1186/s13643-016-0384-4\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWolff RF, Moons KGM, Riley RD et al (2019) PROBAST: A tool to assess the risk of bias and applicability of prediction model studies. Ann Intern Med 170(1):51\u0026ndash;58. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.7326/M18-1376\u003c/span\u003e\u003cspan address=\"10.7326/M18-1376\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBitew M, Desalegn T, Demissie TB, Belayneh A, Endale M, Eswaramoorthy R (2021) Pharmacokinetics and drug-likeness of antidiabetic flavonoids: Molecular docking and DFT study. PLoS ONE 16(12):e0260853. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1371/journal.pone.0260853\u003c/span\u003e\u003cspan address=\"10.1371/journal.pone.0260853\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eNeese F (2018) Software update: The ORCA program system, version 4.0. WIREs Comput Mol Sci 8(1):e1327. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1002/wcms.1327\u003c/span\u003e\u003cspan address=\"10.1002/wcms.1327\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZhao Y, Truhlar DG (2008) The M06 suite of density functionals for main group thermochemistry, thermochemical kinetics, noncovalent interactions, excited states, and transition elements. Theor Chem Acc 120:215\u0026ndash;241. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1007/s00214-007-0310-x\u003c/span\u003e\u003cspan address=\"10.1007/s00214-007-0310-x\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eClark T, Chandrasekhar J, Spitznagel GW, Schleyer PvR (1983) Efficient diffuse function-augmented basis sets for anion calculations. III. The 3\u0026ndash;21\u0026thinsp;+\u0026thinsp;G basis set for first-row elements, Li\u0026ndash;F. J Comput Chem 4(3):294\u0026ndash;301. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1002/jcc.540040303\u003c/span\u003e\u003cspan address=\"10.1002/jcc.540040303\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eTrott O, Olson AJ (2010) AutoDock Vina: improving the speed and accuracy of docking with a new scoring function, efficient optimization, and multithreading. J Comput Chem 31(2):455\u0026ndash;461. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1002/jcc.21334\u003c/span\u003e\u003cspan address=\"10.1002/jcc.21334\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAbraham MJ, Murtola T, Schulz R et al (2015) GROMACS: High performance molecular simulations through multi-level parallelism from laptops to supercomputers. SoftwareX 1\u0026ndash;2:19\u0026ndash;25. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.softx.2015.06.001\u003c/span\u003e\u003cspan address=\"10.1016/j.softx.2015.06.001\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDaina A, Michielin O, Zoete V (2017) SwissADME: a free web tool to evaluate pharmacokinetics, drug-likeness and medicinal chemistry friendliness of small molecules. Sci Rep 7:42717. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1038/srep42717\u003c/span\u003e\u003cspan address=\"10.1038/srep42717\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eButler KT, Davies DW, Cartwright H, Isayev O, Walsh A (2018) Machine learning for molecular and materials science. Nature 559(7715):547\u0026ndash;555. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1038/s41586-018-0337-2\u003c/span\u003e\u003cspan address=\"10.1038/s41586-018-0337-2\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWilkinson MD, Dumontier M, Aalbersberg IJ et al (2016) The FAIR Guiding Principles for scientific data management and stewardship. Sci Data 3:160018. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1038/sdata.2016.18\u003c/span\u003e\u003cspan address=\"10.1038/sdata.2016.18\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGrimme S (2006) Semiempirical GGA-type density functional constructed with a long-range dispersion correction. J Comput Chem 27(15):1787\u0026ndash;1799. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1002/jcc.20495\u003c/span\u003e\u003cspan address=\"10.1002/jcc.20495\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMeng X-Y, Zhang H-X, Mezei M, Cui M (2011) Molecular docking: a powerful approach for structure-based drug discovery. Curr Comput -Aided Drug Des 7(2):146\u0026ndash;157. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.2174/157340911795677602\u003c/span\u003e\u003cspan address=\"10.2174/157340911795677602\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMishra SS, Rai SK, Raghuwanshi HS et al (2021) Computational investigation of potential inhibitors of novel coronavirus 2019 through structure-based virtual screening, molecular dynamics and density functional theory studies. J Biomol Struct Dyn 39(12):4449\u0026ndash;4461. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1080/07391102.2020.1791957\u003c/span\u003e\u003cspan address=\"10.1080/07391102.2020.1791957\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHariharan PC, Pople JA (1973) The influence of polarization functions on molecular orbital hydrogenation energies. Theor Chim Acta 28(3):213\u0026ndash;222. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1007/BF00533485\u003c/span\u003e\u003cspan address=\"10.1007/BF00533485\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eTomasi J, Mennucci B, Cammi R (2005) Quantum mechanical continuum solvation models. Chem Rev 105(8):2999\u0026ndash;3093. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1021/cr9904009\u003c/span\u003e\u003cspan address=\"10.1021/cr9904009\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGoerigk L, Grimme S (2011) Efficient and accurate double-hybrid-meta-GGA density functionals\u0026mdash;evaluation with the extended GMTKN30 database for general main group thermochemistry, kinetics, and noncovalent interactions. J Chem Theory Comput 7(2):291\u0026ndash;309. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1021/ct100466k\u003c/span\u003e\u003cspan address=\"10.1021/ct100466k\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePal R, Patra SG, Chattaraj PK (2022) Quantitative structure\u0026ndash;toxicity relationship in bioactive molecules from a conceptual DFT perspective. Pharmaceuticals 15(11):1383. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3390/ph15111383\u003c/span\u003e\u003cspan address=\"10.3390/ph15111383\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHughes JP, Rees S, Kalindjian SB, Philpott KL (2011) Principles of early drug discovery. Br J Pharmacol 162(6):1239\u0026ndash;1249. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1111/j.1476-5381.2010.01127.x\u003c/span\u003e\u003cspan address=\"10.1111/j.1476-5381.2010.01127.x\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGoldsmith BR, Esterhuizen J, Liu J-X, Bartel CJ, Sutton C (2018) Machine learning for heterogeneous catalyst design and discovery. AIChE J 64(7):2311\u0026ndash;2323. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1002/aic.16198\u003c/span\u003e\u003cspan address=\"10.1002/aic.16198\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLeach AR (2001) \u003cem\u003eMolecular Modelling: Principles and Applications\u003c/em\u003e, 2nd ed. Harlow: Prentice Hall, ISBN: 978-0582382107\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLane TJ, Shukla D, Beauchamp KA, Pande VS (2013) To milliseconds and beyond: challenges in the simulation of protein folding. Curr Opin Struct Biol 23(1):58\u0026ndash;65. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.sbi.2012.11.002\u003c/span\u003e\u003cspan address=\"10.1016/j.sbi.2012.11.002\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGrimme S (2006) Semiempirical hybrid density functional with perturbative second-order correlation. J Chem Phys 124(3):034108. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1063/1.2148954\u003c/span\u003e\u003cspan address=\"10.1063/1.2148954\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCao Y, Romero J, Olson JP et al (2019) Quantum chemistry in the age of quantum computing. Chem Rev 119(19):1085610915. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1021/acs.chemrev.8b00803\u003c/span\u003e\u003cspan address=\"10.1021/acs.chemrev.8b00803\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCaramelli D, Granda S, Cambi\u0026eacute; L et al (2021) Discovering new chemistry with an autonomous robotic platform. ACS Cent Sci 7(11):1821\u0026ndash;1830. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1021/acscentsci.1c00435\u003c/span\u003e\u003cspan address=\"10.1021/acscentsci.1c00435\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZhang H, Juraskova V, Duarte F (2024) Modelling chemical processes in explicit solvents with machine learning potentials. Nat Commun 15(1):6114. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1038/s41467-024-50418-6\u003c/span\u003e\u003cspan address=\"10.1038/s41467-024-50418-6\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":true,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"journal-of-computer-aided-molecular-design","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"jcam","sideBox":"Learn more about [Journal of Computer-Aided Molecular Design](http://link.springer.com/journal/10822)","snPcode":"10822","submissionUrl":"https://submission.nature.com/new-submission/10822/3","title":"Journal of Computer-Aided Molecular Design","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"em","reportingPortfolio":"Springer Hybrid","inReviewEnabled":true,"inReviewRevisionsEnabled":false},"keywords":"Density Functional Theory, computer-aided molecular design, drug discovery, molecular dynamics, ADMET, PRISMA","lastPublishedDoi":"10.21203/rs.3.rs-9151448/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-9151448/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eDensity Functional Theory (DFT) is becoming an increasingly valuable tool in computer-aided molecular design because it provides quantum-level insight that can strengthen practical drug discovery workflows. Beyond its traditional role in electronic structure analysis, DFT is now widely used to support reactivity assessment, binding interpretation, and molecular descriptor generation within integrated in silico strategies. This systematic review examines how DFT has been applied in recent drug discovery research, with particular attention to methodological trends, therapeutic targets, and its integration with other computational approaches. Following PRISMA 2020 guidelines, studies published between January 2021 and February 2025 were identified through searches of Google Scholar, PubMed/MEDLINE, Web of Science, Scopus, and IEEE Xplore. Of 815 records screened, 81 studies met the inclusion criteria. B3LYP and 6-31G(d,p) remained the most commonly used functional and basis set, appearing in 67.9% and 71.6% of studies, respectively. DFT was most often combined with molecular docking (76.5%), ADMET prediction (55.6%), and molecular dynamics (34.6%), highlighting its growing role in multi-method drug design workflows. The main application areas were infectious diseases, cancer, and metabolic disorders. Overall, the evidence suggests that DFT is no longer used primarily as a stand-alone quantum chemical method, but rather as an integrated component of modern molecular design pipelines that supports mechanistic understanding and rational compound optimization. Future advances will likely come from closer integration with machine learning, better treatment of conformational flexibility, and more realistic solvent-aware simulations.\u003c/p\u003e","manuscriptTitle":"Density Functional Theory in Computer-Aided Drug Design: A Systematic Review of Applications, Methodological Trends, and Integration with Molecular Modeling Workflows","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2026-03-20 19:03:22","doi":"10.21203/rs.3.rs-9151448/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"reviewersInvited","content":"","date":"2026-03-30T15:04:38+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2026-03-30T15:02:50+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2026-03-18T12:16:21+00:00","index":"","fulltext":""},{"type":"submitted","content":"Journal of Computer-Aided Molecular Design","date":"2026-03-17T17:10:37+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"journal-of-computer-aided-molecular-design","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"jcam","sideBox":"Learn more about [Journal of Computer-Aided Molecular Design](http://link.springer.com/journal/10822)","snPcode":"10822","submissionUrl":"https://submission.nature.com/new-submission/10822/3","title":"Journal of Computer-Aided Molecular Design","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"em","reportingPortfolio":"Springer Hybrid","inReviewEnabled":true,"inReviewRevisionsEnabled":false}}],"origin":"","ownerIdentity":"45b89dd8-2cf6-490b-a3cb-a44f31cf8b09","owner":[],"postedDate":"March 20th, 2026","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"under-review","subjectAreas":[],"tags":[],"updatedAt":"2026-03-30T15:09:48+00:00","versionOfRecord":[],"versionCreatedAt":"2026-03-20 19:03:22","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-9151448","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-9151448","identity":"rs-9151448","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}