Moremi Bio Agent: Leveraging Agentic Large Language Model for the Discovery of Broad-Spectrum Antibiotics for Enterobacteriaceae

Moremi Bio Agent: Leveraging Agentic Large Language Model for the Discovery of Broad-Spectrum Antibiotics for Enterobacteriaceae | bioRxiv /* */ /* */ <!-- <!-- /*! * yepnope1.5.4 * (c) WTFPL, GPLv2 */ (function(a,b,c){function d(a){return"[object Function]"==o.call(a)}function e(a){return"string"==typeof a}function f(){}function g(a){return!a||"loaded"==a||"complete"==a||"uninitialized"==a}function h(){var a=p.shift();q=1,a?a.t?m(function(){("c"==a.t?B.injectCss:B.injectJs)(a.s,0,a.a,a.x,a.e,1)},0):(a(),h()):q=0}function i(a,c,d,e,f,i,j){function k(b){if(!o&&g(l.readyState)&&(u.r=o=1,!q&&h(),l.onload=l.onreadystatechange=null,b)){"img"!=a&&m(function(){t.removeChild(l)},50);for(var d in y[c])y[c].hasOwnProperty(d)&&y[c][d].onload()}}var j=j||B.errorTimeout,l=b.createElement(a),o=0,r=0,u={t:d,s:c,e:f,a:i,x:j};1===y[c]&&(r=1,y[c]=[]),"object"==a?l.data=c:(l.src=c,l.type=a),l.width=l.height="0",l.onerror=l.onload=l.onreadystatechange=function(){k.call(this,r)},p.splice(e,0,u),"img"!=a&&(r||2===y[c]?(t.insertBefore(l,s?null:n),m(k,j)):y[c].push(l))}function j(a,b,c,d,f){return q=0,b=b||"j",e(a)?i("c"==b?v:u,a,b,this.i++,c,d,f):(p.splice(this.i++,0,a),1==p.length&&h()),this}function k(){var a=B;return a.loader={load:j,i:0},a}var l=b.documentElement,m=a.setTimeout,n=b.getElementsByTagName("script")[0],o={}.toString,p=[],q=0,r="MozAppearance"in l.style,s=r&&!!b.createRange().compareNode,t=s?l:n.parentNode,l=a.opera&&"[object Opera]"==o.call(a.opera),l=!!b.attachEvent&&!l,u=r?"object":l?"script":"img",v=l?"script":u,w=Array.isArray||function(a){return"[object Array]"==o.call(a)},x=[],y={},z={timeout:function(a,b){return b.length&&(a.timeout=b[0]),a}},A,B;B=function(a){function b(a){var a=a.split("!"),b=x.length,c=a.pop(),d=a.length,c={url:c,origUrl:c,prefixes:a},e,f,g;for(f=0;f<d;f++)g=a[f].split("="),(e=z[g.shift()])&&(c=e(c,g));for(f=0;f<b;f++)c=x[f](c);return c}function g(a,e,f,g,h){var i=b(a),j=i.autoCallback;i.url.split(".").pop().split("?").shift(),i.bypass||(e&&(e=d(e)?e:e[a]||e[g]||e[a.split("/").pop().split("?")[0]]),i.instead?i.instead(a,e,f,g,h):(y[i.url]?i.noexec=!0:y[i.url]=1,f.load(i.url,i.forceCSS||!i.forceJS&&"css"==i.url.split(".").pop().split("?").shift()?"c":c,i.noexec,i.attrs,i.timeout),(d(e)||d(j))&&f.load(function(){k(),e&&e(i.origUrl,h,g),j&&j(i.origUrl,h,g),y[i.url]=2})))}function h(a,b){function c(a,c){if(a){if(e(a))c||(j=function(){var a=[].slice.call(arguments);k.apply(this,a),l()}),g(a,j,b,0,h);else if(Object(a)===a)for(n in m=function(){var b=0,c;for(c in a)a.hasOwnProperty(c)&&b++;return b}(),a)a.hasOwnProperty(n)&&(!c&&!--m&&(d(j)?j=function(){var a=[].slice.call(arguments);k.apply(this,a),l()}:j[n]=function(a){return function(){var b=[].slice.call(arguments);a&&a.apply(this,b),l()}}(k[n])),g(a[n],j,b,n,h))}else!c&&l()}var h=!!a.test,i=a.load||a.both,j=a.callback||f,k=j,l=a.complete||f,m,n;c(h?a.yep:a.nope,!!i),i&&c(i)}var i,j,l=this.yepnope.loader;if(e(a))g(a,0,l,0);else if(w(a))for(i=0;i (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];var j=d.createElement(s);var dl=l!='dataLayer'?'&l='+l:'';j.src='//www.googletagmanager.com/gtm.js?id='+i+dl;j.type='text/javascript';j.async=true;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-M677548'); Skip to main content Home About Submit ALERTS / RSS Search for this keyword Advanced Search New Results Moremi Bio Agent: Leveraging Agentic Large Language Model for the Discovery of Broad-Spectrum Antibiotics for Enterobacteriaceae Gertrude Hattoh , Jeremiah Ayensu , Nyarko Prince Ofori , Solomon Eshun , Joshua Ntow Opare-Boateng , Osman Tanko , Darlington Akogo doi: https://doi.org/10.1101/2025.08.21.671656 Gertrude Hattoh 1 MinoHealth AI Labs , Accra, Ghana Find this author on Google Scholar Find this author on PubMed Search for this author on this site Jeremiah Ayensu 1 MinoHealth AI Labs , Accra, Ghana Find this author on Google Scholar Find this author on PubMed Search for this author on this site Nyarko Prince Ofori 1 MinoHealth AI Labs , Accra, Ghana Find this author on Google Scholar Find this author on PubMed Search for this author on this site Solomon Eshun 1 MinoHealth AI Labs , Accra, Ghana Find this author on Google Scholar Find this author on PubMed Search for this author on this site Joshua Ntow Opare-Boateng 1 MinoHealth AI Labs , Accra, Ghana Find this author on Google Scholar Find this author on PubMed Search for this author on this site Osman Tanko 1 MinoHealth AI Labs , Accra, Ghana Find this author on Google Scholar Find this author on PubMed Search for this author on this site Darlington Akogo 1 MinoHealth AI Labs , Accra, Ghana Find this author on Google Scholar Find this author on PubMed Search for this author on this site For correspondence: ahialed{at}gmail.com Abstract Full Text Info/History Metrics Preview PDF Abstract Antimicrobial resistance (AMR) is a pressing global health crisis, exacerbated by a stagnating antibiotic discovery pipeline and the emergence of multidrug-resistant pathogens such as Klebsiella pneumoniae . We hypothesize that dual-target strategies may offer a more robust means to overcome AMR by reducing the likelihood of resistance development compared to single-target approaches. In response, we leveraged Moremi Bio Agent, an agentic large language model (LLM) for the autonomous design, in silico validation, and prioritization of broad-spectrum antibiotics targeting Enterobacteriaceae. Using our proposed dual-target strategy, we generated and evaluated 1,002 candidate molecules predicted to simultaneously inhibit the FabI enzyme and the AcrAB–TolC efflux pump—two key resistance mechanisms in Gram-negative bacteria. Our fully autonomous pipeline integrated compound generation, molecular docking, pharmacodynamics/pharmacokinetic predictions & toxicity profiling, and molecule-ranking based on ADMET and drug-likeness properties. Out of 1,002 molecules generated, 774 passed preliminary ADMET benchmarks, with majority of the 60 top-performing candidates (score ≥ 0.8) showing favorable drug-likeness, minimal toxicity. 391 of the compounds exhibited moderate binding interaction to both targets. This study demonstrates the feasibility of AI-driven antibiotic discovery and lays the foundation for future experimental validation to address AMR. 1 Introduction Antimicrobial Resistance (AMR) is a significant global health threat, recognized by the World Health Organization (WHO) as one of the top global public health threats facing humanity World Health Organization (2023) . This has been a critical challenge to healthcare and antimicrobial research, largely due to overuse and misuse of antimicrobial drugs Salam et al. (2023) . The Global Research on Antimicrobial Resistance, GRAM reports that each year, AMR causes significant deaths worldwide; approximately 1.27 million deaths were recorded in 2019 Kim et al. (2022) and a projected economic burden of $100 trillion USD by 2050 if no effective new antibiotics are developed O’neill (2014) . The global fight against AMR faces significant challenges due to a stagnant antibiotic pipeline Årdal et al. (2019). Despite urgent needs, there is a critical shortfall in new antibiotics reaching clinical development. This is largely attributed to high research and development (R&D) costs Schlander et al. (2021) , which deter investment in antibiotic innovation. The current pipeline is described as inadequate to address the mounting threat of antibiotic resistance, leaving patients vulnerable to increasingly resistant bacterial infections. Consequently, the limited discovery of novel antibiotics has significantly hindered efforts to combat AMR Laxminarayan et al. (2013) , allowing drug-resistant pathogens to proliferate unchecked. This necessitates the need for new therapeutic approaches Vaja et al. (2024) . Amidst these challenges, Artificial Intelligence (AI), particularly Large Language Models (LLMs), has emerged as a transformative force in biotechnology, revolutionizing drug discovery, protein engineering, and antibody development Xiang et al. (2024) . In the context of antimicrobial resistance (AMR), AI-driven approaches have shown substantial promise across key domains, including discovery and design, diagnostic support, and data integration—each contributing to accelerated progress against AMR Yoo et al. (2024) . Notable advancements include the use of generative models to design antibiotics targeting Acinetobacter baumannii Swanson et al. (2024a) , the development of novel antimicrobial peptides Wang et al. (2024) , molecular optimization, and the acceleration of early-stage drug discoveryTorres et al. (2024), Delmas et al. (2025) . Amidst global efforts to tackle antimicrobial resistance (AMR), Minohealth AI Labs is leading innovation in Africa and beyond. Our AI model, Moremi Bio Agent—already proven in autonomously designing and validating novel antibodies against malaria and small molecules for diverse diseases Akogo et al. (2025) , MinohealthAILabs (2025) —has now been deployed to generate an initial library of 1,002 novel broad-spectrum antibiotic candidates targeting Enterobacteriaceae, using Klebsiella species as the starting point. This foundational effort aims to address critical gaps in antibiotic discovery research. Our approach not only accelerates the rapid generation of novel antibiotic scaffolds but also explores the potential of a dual-target strategy—a hypothesis we propose may offer a more robust means to overcome AMR by reducing the likelihood of resistance development. Through comprehensive in silico high-throughput design, validation, and iterative optimization, we will refine and prioritize the most promising candidates, advancing them through successive stages, from computational refinement to possible design enhancements and experimental testing. This initiative sets the stage for a scalable and high-throughput framework designed to meet the AMR challenge where traditional methods have struggled. 2 Literature Review 2.1 Background Antimicrobial resistance (AMR) is an escalating global health crisis, contributing to substantial morbidity, mortality, and economic burden worldwide 201 (2019). Infections caused by multidrug-resistant pathogens—especially members of the Enterobacteriaceae family such as Klebsiella spp., pose a severe threat to public health Podschun and Ullmann (1998) . Klebsiella pneumoniae , in particular, is notorious for its ability to acquire resistance mechanisms, rendering many conventional antibiotics ineffective and complicating treatment regimens for critically ill patients Li et al. (2023 , 2024 ). Traditional antibiotic discovery methods, including high-throughput screening and conventional computational approaches, have been hampered by extensive timeframes, high costs, and limited success rates. These methods often fail to explore the vast chemical space comprehensively, thereby missing novel chemical scaffolds with unique multi-target mechanisms that are critical for circumventing bacterial resistance Terekhov et al. (2018) , Silver (2011) . Recent landmark discoveries—such as the identification of Halicin through artificial intelligence—have demonstrated the potential of machine learning (ML) techniques to uncover new antibiotics. However, these advancements also underscore the ongoing need for innovative approaches to rapidly identify and validate antibiotic candidates with novel mechanisms of action, which can address the growing challenge of antimicrobial resistance. Possibly by integrating multi-target design strategies into their generative frameworks, LLMs can explore vast chemical spaces and propose novel antibiotic structures in a matter of minutes—a stark contrast to the prolonged timelines of traditional experimental methods. In this study, we present an innovative approach that harnesses LLMs to accelerate the discovery of novel broad-spectrum antibiotics specifically targeted against members of the Enterobacteriacea starting with Klebsiella spp. Our method proposes a dual-target strategy to design candidate molecules that are predicted to interact with two essential bacterial pathways Gray and Wenzel (2020) . These candidate molecules are then subjected to rigorous in silico assessments, including evaluations of their Absorption, Distribution, Metabolism, Excretion and Toxicity (ADMET) properties, stability, and binding affinities with proposed bacterial targets. This integrated computational pipeline not only significantly reduces the time and cost associated with antibiotic development but also enhances the likelihood of identifying candidates with robust efficacy against members of Enterobacteriaceae. 2.2 The Global Urgency for Gram-Negative Antibiotics According to the WHO Priority Pathogens List, Gram-negative bacteria represent one of the fore-most challenges in combating antimicrobial resistance Moja et al. (2017). In combating AMR, it is essential to prioritize Gram-negative bacteria since they exhibit greater resistance than their Gram-positive counterparts—owing to their inherent defense mechanisms, intricate cell architecture, and the increasing burden of hospital-acquired infections Melander et al. (2022) . The WHO’s priority list highlights critical Gram negative pathogens including the family Enterobacteriaceae as those among the ‘critical’ class Moja et al. (2017). Many reports indicate that resistance to key antibiotics such as third-generation cephalosporins and carbapenems is rising within this family. For instance, WHO data and regional surveillance efforts have documented that in some countries, up to 42% of E. coli isolates are resistant to third-generation cephalosporins, while similar high resistance levels are seen among K. pneumoniae . These trends contribute heavily to the morbidity and mortality from bloodstream and urinary tract infections, among other conditions G et al. (2022). Gram-negative pathogens owe much of their antimicrobial resistance to their complex cell envelope architecture: a dense lipopolysaccharide (LPS) outer membrane that restricts drug entry, porin modifications that further limit influx, robust efflux pump systems that expel antibiotics, and a suite of β -lactamases and other enzymes that degrade drugs before they can act. These intertwined barriers underscore the urgent need for innovative therapeutic strategies that can breach or bypass them. While approaches such as phage therapy and immunomodulation are gaining traction Bhandari and Suresh (2022) , World Health Organization (2023) , Alipour-Khezri et al. (2024) , antibiotics remain indispensable, even as their efficacy wanes. To invigorate the antibiotic arsenal, we propose coupling LLM-powered exploration of vast chemical space with a polypharmacology paradigm to design single molecules that engage two bacterial targets. We hypothesize that such multi-target (dual-action) agents will significantly lower the chance of resistance emergence by striking two critical pathways at once Feng et al. (2023) . Embedding this concept within a structured pipeline—spanning in silico discovery, rigorous computational validation, and iterative optimization—offers a path toward next-generation antibiotics capable of overcoming Gram-negative defenses where traditional monotherapies have repeatedly fallen short Khabthani et al. (2021) . 2.3 Significance of Targeting Enterobacteriaceae and Klebsiella species Enterobacteriaceae represent the largest and most diverse family of Gram-negative bacteria Oliveira and Reygaert (2023) . Among these, Escherichia coli and Klebsiella pneumoniae are consistently identified as the leading pathogens in Hospital-acquired infections (HIAs) Irek et al. (2018) , Saleem et al. (2019). The 2024 WHO Bacterial Priority Pathogens List (BPPL) classifies carbapenemresistant Enterobacterales—including carbapenem-resistant K. pneumoniae —in its highest priority tier, reflecting their urgent need for novel therapeutics Sati et al. (2025) . Klebsiella spp. are ubiquitous and cause a wide range of infections Bengoechea and Pessoa (2018) which includes bloodstream infections, urinary tract infections, pneumonia and other life threatening conditions Xu et al. (2018) , Gonzalez-Ferrer et al. (2021) , Chang et al. (2021) . The pathogenic success of Klebsiella spp. and related Enterobacteriaceae stems from an arsenal of resistance mechanisms: production of β –lactamases and carbapenemases that inactivate antibiotics, modification of drug targets, loss or mutation of outer-membrane porins, upregulation of efflux pumps, and formation of biofilms that impede antimicrobial penetration Paczosa and Mecsas (2016) , Wang et al. (2020) . Klebsiella spp, more specially K. pneumoniae is an opportunistic pathogen, which causes around 6-10% of nosocomial infections worldwide Navon-Venezia et al. (2017) , Asri et al. (2021), Shah et al. (2025) . Its role as a major public health concern is heightened by its ability to rapidly develop resistance to multiple antibiotic classes including carbapenem 201 (2019), placing it at the top of the World Health Organization’s 2024 Bacterial Priority Pathogens List. In the United States, the CDC’s 2019 report on Antibiotic Resistance Threats specifically lists Klebsiella as one of the most pressing concerns for antibiotic resistance 201 (2019). K. pneumoniae is recognized as an important multidrug-resistant (MDR) pathogen, with increasing prevalence of resistance to first-line and last-resort antibiotics complicating clinical therapy Li et al. (2022 , 2023 ). Klebsiella spp provides a physiologically relevant model for screening compounds against traits common to Enterobacteriaceae; outer-membrane permeability barriers, resistance, virulence and β -lactamase activity Li et al. (2025) . Also, using Klebsiella spp. as a starting point aligns with global research initiatives (E.g Grand Challenges in Gram-negative Antibiotic Discovery) which designates Klebsiella spp. for early-phase drug discovery to foster collaboration and accelerate the development of broad-spectrum agents against Enterobacteriaceae. 1 Therefore, a thorough understanding of Klebsiella resistance and pathogenicity is critical for guiding the discovery of novel antibiotics with activity across the Enterobacteriaceae family and for addressing the global antimicrobial resistance crisis. 2.4 Fabl and AcrAB-TolC efflux pump Our goal is to leverage Moremi Bio to autonomously generate and in silico validate one thousand and two (1,002) novel compounds; adopting a polypharmacology strategy, these candidates were designed to inhibit the FabI enzyme while simultaneously blocking the AcrAB-TolC pump in a high-throughput manner. FaBI are enoyl-ACP (Acyl Carrier Protein) reductases critical in the pathway II fatty acid biosynthesis (FAS II) in most bacteria Lu and Tonge (2008) . Fatty acids play essential roles in the bacteria from use for membrane synthesis essential for their survival through to contribution of virulence and pathogenesis Waters and Eijkelkamp (2024) . FabI occupies a pivotal, rate-determining position in the bacterial FAS II pathway: as the terminal enoyl-ACP reductase, it leverages a highly favorable reduction to pull the elongation cycle to completion and regulate flux into membrane phospholipid assembly Parsons and Rock (2013). FaBI has been considered a significant therapeutic target for antibacterial drug discovery, highly conserved components in many pathogens and are non-homologous to human targets; targeting the Fabl enzyme not only inhibits bacterial infections but also leads to a broad spectrum antimicrobial activity Rana et al. (2020) . While bacteria can incorporate extracellular fatty acids, exogenous lipids fail to fully compensate for the essential FAS-II derived intermediates and can not bypass the need for de novo fatty acid synthesis; ultimately blocking fatty synthesis remains critical and an effective antimicrobial strategy Yao et al. (2015) , Rana et al. (2020) , Biswas et al. (2023) . [Fabl structure from the PDB database with PDB 8JTP]. AcrAB-TolC efflux pump is a tripartite complex composed of the AcrA, AcrB and TolC proteins of the Resistance-nodulation-division (RND)-family efflux pumps; a major mechanism by which Klebsiella resists multiple classes of antibiotics Wand et al. (2022) . TolC penetrates the outer membrane bacteria lipid bilayer, AcrA is positioned within the layer and forms the bridge between the AcrB on the inner to the TolC on the outer cell membrane Wang et al. (2017) , Chen et al. (2020) [Structure from the PDB Database with id = 5O66]. The AcrAB-TolC efflux pump has various roles which includes multidrug resistance, biofilm formation and virulence, and stress adaptation Padilla et al. (2009) , Jang (2023) . Targeting both Fabl and the AcrAB-TolC efflux pump simultaneously presents a promising polypharmacological strategy to combat AMR. Fabl inhibition disrupts bacterial membrane biosynthesis by blocking the enoyl-ACP reductase step in fatty acid synthesis Mehboob et al. (2012) and AcrAB-TolC inhibition prevents antibiotics efflux hence increasing the intracellular drug concentrations Alenazy (2022) ; which intend create a lethal accumulation effect. 3 Methodology This project employed the use of LLM-based approach to mine the literature, identify relevant antibacterial targets, and designed novel antibiotics against Klebsiella spp. Using our model, Moremi Bio, we adopted a polypharmacology strategy to autonomously generate and in silico evaluate one thousand and two (1,002) novel antibiotic candidates. These candidates were designed to inhibit the FabI enzyme while simultaneously blocking the AcrAB-TolC efflux pump in a high-throughput manner. The compound generation and validation pipeline, which is the pre-developed agentic model capable of autonomously executing these tasks without human in-the-loop was leveraged. Moremi Bio Agent integrated with advanced external industry-standard tools to validate generated antibiotics. The validation process involved molecular docking studies to assess binding affinity with the targets and in silico predictions of the Pharmacodynamics/Pharmacokinetics (PD/PK). The assessment also evaluated toxicity (LD 50 ), off-target interactions (hERG), and stability to ensure a comprehensive evaluation of the antibiotic candidates. A ranking system in the pipeline assigned ranks to all molecules, facilitating the easy identification of lead compounds based on their ADMET properties and other relevant properties [See Appendix A]. An overview of the method is presented in an image below, figure 1 . Download figure Open in new tab Figure 1: Overview of Moremi Bio Agent method A framework of the computational design, agentic pipeline with relevant components and verification of novel compounds. 4 Results This research applied an AI-driven approach to swiftly design unique antibiotic molecules with dual-target inhibition to Fabl and AcrAB-TolC efflux pump, with validated PD/PK properties, and accessed on several critical criteria which includes Binding affinity assessment; providing a promising foundation for future experimental validation against AMR pathogens. Moremi Bio generated a total of 1002 molecules of which 228 failed on several benchmarks. Out of the 774 candidate molecules that passed preliminary benchmarks, 545 (approximately 70.4%) had molecular weights below 700 Da. 4.1 PD/PK Properties Performing in silico assessment of relevant drug-likeness of generated molecules significantly raises the probability in identifying leads and areas for further optimization before further in vitro-in vivo studies Dhoble et al. (2025) . PD/PK of molecules were assessed using the ADMETAI Swanson et al. (2024b) ; these included Absorption, Distribution, Metabolism, Excretion, and Toxicity (ADMET) predictions. Several metrics are inclusive to obtain each category of the ADMET prediction of the novel compounds as predicted through the ADMETAI [See Appendix A]. The figures below summarizes the descriptive and distributive analysis of the generated compounds. This presents a broad overview to the total landscape distribution of all generated molecules across the ADMET properties. Each subplot shows a histogram overlaid with a kernel density estimate (KDE), illustrating both frequency counts and the underlying distribution shape. The model predicted high absorption (scores ≥ 0.5) for 72% of candidates. The distribution score (subplot2) shows a prominent peak around 0.45 indicating that nearly half of the compounds have moderate tissue distribution and a gentle right shows that fewer molecules achieved scores ≥ 0.8. The Metabolic scores show a pronounced right skew revealing that majority of the compounds were metabolically stable. The Excretion score unlike other metrics shows a bimodal distribution. The toxicity distribution score, based on predicted hERG inhibition probabilities, indicates that a substantial proportion of the compounds exhibit low predicted hERG inhibition (i.e., low probability of cardiotoxicity), reflected in high toxicity safety scores. This suggests that many of the generated molecules are likely to have a favorable cardiac safety profile, with relatively few exhibiting moderate-to-high risk for cardiotoxic effects. Molecules, each color-coded by its overall score (composite ADMET scores (0-1)) integrating each property listed in A. Those with similar ADMET and drug-likeness profiles cluster together spatially. Regions with lighter colors (higher scores) suggest clusters of molecules with optimal profiles, potentially highlighting promising drug candidates. Conversely, darker regions represent molecules with less desirable properties. This prediction gives an indication that the overall scores of the molecules somewhat correlates with the structural property hence translating into the molecules’ performance. A high-resolution comparison of how our novel antibiotic candidates distribute across key AD-MET properties offer deep insight into their developability profiles. By juxtaposing summary-statistical box plots with full-distribution violin plots, we can both quantify central tendencies and visualize the underlying density shapes. 4.2 Absorption One of the key determinants of oral efficacy for broad-spectrum antibiotics is their ability to be absorbed efficiently through the gut epithelium; which is mostly evaluated in vitro by measuring permeability across cell-based or artificial membrane models (Wang et al., 2016), predicting Human Intestinal Absorption (HIA) along with assessing interactions with P-glycoprotein (Pgp) Kus et al. (2023) is essential early in silico assessment. Caco-2 (Cancer coli) cell line 2 were predicted to range from − 7.578 × log(10 − 6 ) cm/s and − 3.800 × log(10 − 6 ) cm/s, which suggest very high absorption in the gut epithelium Kus et al. (2023) for − 3.800 × log(10 − 6 ) cm/s (i.e., 158 × 10 − 6 cm/s). HIA reached 0.999 (99.9%) exceeding the 80% threshold for high bioavailability (Yan et al., 2008). Parallel Artificial Membrane Permeability Assay (PAMPA) values which indicate the ability of the molecule to diffuse an artificial membrane were as high as 99.7%. These are indicative that the optimal absorption of the generated molecules can possibly reach their therapeutic concentrations in the plasma and significantly raises the probability that they will achieve the systemic exposures required for broad-spectrum activity. 4.3 Distribution Approximately, sixty-three percent ( ≈ 63.00%) of the molecules had optimal Plasma Protein Bound (PPB) values (i.e., 50-90% PPB) Watanabe et al. (2018) . This means an adequate quantity of unbound compounds exist in the plasma to exert antimicrobial effect. 4.4 Metabolism A minimal interaction with the cytochrome p450 (CYP) enzymes is optimal. Predicted probability of cytochrome inhibition measured for some major CYP (Cyp2C9, Cyp2D6 and Cyp3A4) Arakawa et al. (2023) enzymes ranged from 9.427e-05 to 0.980, 2.295e-04 to 0.922 and 7.871e-06 to 0.993 respectively. CYP2C9 inhibition probabilities up to 0.98 indicate high risk for drug–drug interactions in candidates with hepatic metabolism pathways Boiko et al. (2020) . The top-10 ranked candidates showed minimal interaction with the cytochrome p450 enzymes, the mean of the min and max inhibition values across the three CYP inhibition values ranged from o.0016 to 0.1406, suggesting an acceptable range which translates to reduced hepatic elimination, longer in vivo half-lives, and potentially less frequent dosing in humans. 4.5 Excretion Approximately, 37.60% of molecules exhibited Clearance(Cl) less than 5 µ L/min/10 6 cells ( < 15 mL min − 1 kg − 1 ); the large low-Cl fraction suggests that several compounds may achieve sustained plasma exposure ideal for good dosing regimens. The highest clearance rate recorded was 28.462 µ L/min/10 − 6 cells S-loczyńska et al. (2019). Half-life ( t 1 / 2 ) inversely correlates with clearance (Cl), as defined by t 1 / 2 = 0.693×Vd/Cl, demonstrating that half-life is directly proportional to volume of distribution and inversely proportional to clearance Grogan and Preuss (2023) . The observed negative correlation in figure 5 aligns with established pharmacokinetic theory. Compounds with elevated clearance demonstrate shorter half-lives, potentially necessitating more frequent dosing regimens. Conversely, lower clearance values typically yield extended half-lives, which may facilitate less frequent administration schedules and more sustained therapeutic plasma concentrations. Download figure Open in new tab Figure 2: Distribution of compounds profiled access the five ADMET metrics Download figure Open in new tab Figure 3: t-SNE projection of ADMET properties and drug-likeness profiles Download figure Open in new tab Figure 4: Distribution comparison across ADMET properties A : Box and whisker plot of distribution across each ADMET property. B : Violin plot showing comparison across ADMET properties. From the distribution in A and B , most candidates excel at Metabolism and equally exhibit good Absorption, Distribution, Excretion, and Toxicity. Download figure Open in new tab Figure 5: Relationship between half-life and clearance. The substantial data scatter observed suggest significant variability in the clearance-half-life relationship. 4.6 Toxicity and off target interaction Lethal Dose 50 (LD 50 ) and the human Ether-ía-go-go-Related Gene (hERG) inhibition were measured to define the potential toxicity and off-target interaction of the generated molecules; as early assessment of toxicity through in silico means is desirable to ascertain the candidacy for the various generated molecules. hERG is a critical indicator of cardiotoxicity as it encodes the Kv11.1 protein responsible for electrical activity in the heart Vandenberg et al. (2012) . The min and max probability of herG inhibition values ranged from 5.393 × 10 − 4 to 9.813 × 10 − 1 . This gives an indication that some molecules have high potential to cause cardiac side effects Garrido et al. (2020) .Yet, the top-ten ranked molecules 1.799 × 10 − 3 and 3.429 × 10 − 1 (values are closer to zero) indicating a very low probability for hERG inhibition hence suggesting little to no potential cardiac side effects. LD 50 values range from 1.213 to 4.593 in log(1/(mol/kg)). Table 1 below gives a comprehensive information on the LD 50 values of the first ten ranked candidates. View this table: View inline View popup Download powerpoint Table 1: Toxicity and molecular properties of top-10 compounds. Chemical structure of compounds also influence drug-likeness and properties of drugs Mao et al. (2016) , Jia et al. (2019) . Lipinski’s rule of 5 is critical to estimate the bioavailability of a compound based on their chemical structure Benet et al. (2016) , Daina et al. (2017) . In figure 7 , plot 1 (box plot, lipinski Violations) indicates the number of molecules with their corresponding number of violations to Lipinski’s rule. A great proportion of the molecules had zero (0/ no violations). As illustrated in plot 2 (Bioavailability Score Dist) the QED values for assessing the desirability of drugs Bickerton et al. (2012) cluster between 0.4–0.6, suggesting moderate drug-likeness, with a minority in the undesirable range (0.1–0.2). Download figure Open in new tab Figure 6: Strength of interaction between overall scores of all 8 category Download figure Open in new tab Figure 7: Drug-likeness and Bioavailability Property Distribution of molecules Subplots: 1 Lipinski violations 2 Bioavailability Property Dist 3 (Quantitative Estimate of drug-likeness) QED 4 Synthetic Accessibility 5 LogP Distribution 6 TPSA Distribution Considering the first sixty ranked (top-60) molecules based on their overall scores, the majority fall in an acceptable range of ≥ 0.8. This threshold of ≥ 0.8 is significant because molecules achieving this score demonstrate superior ADMET properties, making them promising candidates for further experimental studies The graph demonstrates that all 60 top-ranked molecules exhibit desirable ADMET properties and drug-likeness characteristics, with the majority achieving scores ≥ 0.8. See Appendix B for detailed ADMET predictions on top-10 molecules. 4.7 Antibiotics-Target Interaction Binding Affinity prediction provides insights into the strength of interaction between molecules Kastritis and Bonvin (2013) . The Binding free energy (Δ G ) prediction using automation through Autodock VinaEberhardt et al. (2021) showed that a total of 393 molecules exhibited quantifiable bind for Fabl while 391 did for efflux pump. A total of 391 molecules exhibited binding interactions with both targets. The computational analysis of the binding free energy (ΔG) for FabI (mean ΔG = − 7.28 ± 1.13 kcal/mol); and (mean ΔG = − 7.13 ± 0.92 kcal/mol) for AcrAB–TolC pump. The Binding free energy (ΔG) prediction results for the top-ten performing molecules based on the ADMET ranking revealed that nine of ten of these candidates achieved moderate ΔG values against FabI, spanning –7.398kcal/mol to –5.132kcal/mol and AcrAB–TolC pump ranging from – 7.495kcal/mol and –5.505kcal/mol; this is indicative of moderate binding interactions [See Appendix D for detailed binding interaction of top-10 ranked molecules]. In contrast, for molecule rank 4 no binding free-energy estimates were predicted with both targets. The docking poses for AnBkr1 through AnBkr10 within the FabI binding site are depicted in Figure 11 . Download figure Open in new tab Figure 8: Ranked Overall Score relative to threshold ( > = 0.8) for top 60 compounds. Download figure Open in new tab Figure 9: Binding free energy (ΔG) prediction Download figure Open in new tab Figure 10: Heatmap of Binding Free Energies (kcal/mol) Across Dual Targets. Reveals Patterns of Dual Inhibition. This visualization highlights that these molecules have potential dual-target efficacy. Download figure Open in new tab Download figure Open in new tab Figure 11: Binding interaction of top-10 molecules with the Fabl target. Images are arranged in a 3×3 grid. The protein backbone is rendered in cyan, residues defining the binding site are highlighted in magenta, and the bound ligand is shown in limon. More negative binding free energies (ΔG) typically indicate stronger and more favorable interactions between ligands and molecular targets, with values between –7 and –10 kcal/mol often correlating with low micromolar to nanomolar affinities. The mean binding free energy (ΔG) of the novel compounds against FabI is consistent with literature values reported for 1,3,4-thiadiazolyl thiourea derivatives, which exhibited docking scores ranging from –6.56 to –8.67 kcal/mol against FabI George et al. (2014) . Regarding the AcrAB-TolC efflux system, our molecules’ binding free energies fall within a similar energetic range to those of known inhibitors. Recent research has uncovered several pyranopyridine derivatives as potent inhibitors, exhibiting strong binding to the AcrB component of the transporter, with docking scores ranging from –10.168 to –8.427 kcal/mol Mahey et al. (2024) . These findings suggest that the newly designed compounds may exhibit binding interactions comparable to known FabI and AcrB inhibitors, supporting their potential as promising candidates for further antibacterial evaluation. 5 Discussion In 2021, AMR was associated with 4.71 million deaths, predominantly affecting low– and middle-income countries (LMICs) Naghavi et al. (2024) . The World Health Organization has identified carbapenem-resistant Enterobacteriaceae (CRE), including K. pneumoniae, as critical AMR global health threats. The escalating prevalence of multidrug-resistant Gram-negative bacteria, particularly within the Enterobacteriaceae family, underscores the urgent need for innovative therapeutic strategies. This research proposed to use Moremi Bio Agent, an agentic LLM, to accelerate the discovery of broad-spectrum antibiotics against Enterobacteriaceae. By utilizing an agentic LLM, we aimed to rapidly design unique antibiotic molecules with dual-target capabilities, significantly reducing the time required compared to traditional methods. The integration of computational assessments for ADMET properties and binding affinities further streamlines the validation process, minimizing reliance on time-consuming procedures. From the results shown, it indicates that our model designed molecules with promising ADMET properties. In contrast, the synthetic accessibility of molecules requires optimization; as indicated from subplot 4 (Synthetic Accessibility) there is a potential for difficulty in chemical synthesis (SA values ≥ 5) and potential high manufacturability cost for approximately 98% of generated molecules. Our results demonstrate the value of early in silico evaluations in identifying potential ADMET liabilities and prioritizing compounds for downstream experimental validation Roney and Mohd Aluwi (2024). The results provided from this high-throughput screening of our generated molecules, have given us the ideas into which areas are necessary for optimization and molecules that can stand candidacy for further refinements and optimizations. Further research alongside refinement are required to generate promising candidates with low synthetic accessibility scores and that can interact with both targets in high affinity to further solidify our hypothesis of the dual-target mechanism of generated molecules. By harnessing this LLM technology, we aim to uncover innovative chemical scaffolds beyond the reach of traditional methods, and by pursuing a dual-target approach, we seek to overcome antibiotic resistance. 5.1 Limitations All results reported here are derived solely from in silico analyses and therefore demand extensive empirical validation. To establish the true efficacy and safety profile of the generated compounds, follow-up studies must extend beyond computational predictions to include comprehensive assessment of target engagement, antimicrobial potency under physiologically relevant conditions, among others. Moreover, confirmation of any synergistic or additive effects will require functional combination assays under varying conditions to capture kinetic as well as static interactions. Finally, it should be noted that a subset of the screened molecules showed no detectable Fabl and AcrAB-TolC efflux-pump activity, underscoring the importance of further experiments to rule out false negatives and to fully characterize their resistance-modulating potential. 5.2 Innovation and global impact This research addresses a critical gap in antibiotic discovery by integrating LLM approach to rapidly generate novel, dual-target antibiotic candidates against Klebsiella spp with broad-spectrum activity for Enterobacteriaceae. This innovative approach aims to overcome the limitations of traditional methods, which are often time-consuming and costly. Beyond its immediate scientific contributions, this research offers a scalable and cost-effective framework for future antibiotic discovery, ensuring a sustainable pipeline of new therapeutics to mitigate the AMR crisis. By employing advanced AI techniques, the project seeks to expedite the identification of effective antibiotics, mitigate the economic and health burdens associated with AMR and thereby contribute to global efforts in combating AMR. By focusing on AI-driven antibiotic discovery, this research not only addresses the immediate need for new antibiotics but also establishes a framework for integrating AI in drug development. This project exemplifies a pioneering effort to harness AI in addressing one of the most pressing health challenges of our time, with the potential to deliver substantial benefits on a global scale. 6 Conclusion This study demonstrates the potential of AI-driven molecular design to rapidly generate and prioritize novel antibiotic candidates with promising ADMET and drug-likeness profiles. While several compounds exhibited favorable in silico properties and moderate binding to Fabl and AcrAB-TolC efflux-pump, further optimization and experimental validation are needed. The inability to identify strong binders for both targets highlight challenges in the dual-target drug design and underscores the need for improved computational methods or alternative chemotypes. Future work will focus on iterative optimization of the most promising candidates and at most possible synthesis, and in vitro/in vivo testing. Appendices A Ranking of Small Molecules A ranking system integrated into the Moremi Bio Agent pipeline to assign ranks to molecules based on their molecular performance comprehensively. This ranking system ranks the molecules based on eight key property categories, with each category comprising multiple metrics. The table below outlines the key properties and their corresponding metrics that affect the ranking of all generated compounds. View this table: View inline View popup Download powerpoint Table 2: Categories and metrics used in compound ranking. B ADMET characteristics of top-12 molecules The following results provide comprehensive insights into the twelve most promising candidates with broad-spectrum activity against Enterobacteriaceae based on the ranks. The visual below for top-12 promising candidates from (AnBkr1 – AnBkr12). Download figure Open in new tab Figure 12: Comparative ADMET Profiles of Top-12 Ranked compounds C Toxicity LD 50 classification based on Globally Harmonized System (GHS) for chemical classification for top-10 ranked molecules. 2 , 3 View this table: View inline View popup Download powerpoint Table 3: Distribution of molecules across toxicity categories based on LD 50 values. Table 4 summarizes the distribution of predicted LD 50 values for the generated antibiotic candidates. A majority of the compounds (51.16%) fall under Category 4 (Harmful), indicating moderate toxicity levels. Notably, 23.77% of the molecules are classified as Category 3 (Toxic), suggesting a need for structural optimization in this subset. Conversely, 13.05% of the candidates were found to be practically non-toxic, with LD 50 values exceeding 5000 mg/kg, while 10.85% exhibit low toxicity (Category 5). A small fraction (1.16%) demonstrated high toxicity (Category 2), representing high-priority targets for exclusion or redesign. These findings provide an initial safety profile landscape for the antibiotic library and guide further selection for in vivo validation. View this table: View inline View popup Download powerpoint Table 4: Distribution of molecules across toxicity categories. D Binding Free Energy and Potential Dual-target Inhibition of Top-10 Ranked Molecules The figures below present the comparative binding interactions of top-10 candidate compounds across dual targets. Bar chart showing the binding free energies (kcal/mol) of computationally generated antibiotic candidates against two bacterial targets. Each compound is evaluated across both targets to assess dual-inhibition potential. Download figure Open in new tab Figure 13: Distribution comparison across ADMET properties The binding free energies (kcal/mol) of the top 10 candidate compounds on dual targets. View this table: View inline View popup Download powerpoint Table 5: Binding free energies ΔG of top-10 ranked molecules Footnotes ↵ 1 https://gcgh.grandchallenges.org/challenge/innovations-gram-negative-antibiotic-discovery ↵ 2 https://www.ncbi.nlm.nih.gov/books/NBK253967/ ↵ 3 https://webapps.ilo.org/static/english/protection/safework/ghs/ghsfinal/ghsc05.pdf? References 1. Antibiotic resistance threats in the united states , 2019 . 2019. URL https://api.semanticscholar.org/CorpusID:241621099 . 2. ↵ Darlington Ahiale Akogo , Jeremiah Ayensu , Nana Sam , Gertrude Hattoh , Prince Nyarko , Solomon Eshun , Mohammed Alhasan , Henrietta E. Mensah-Brown , and Peter Quashie . Moremi bio agent: Application of a foundation model and end-to-end automation in the design and validation of monoclonal antibodies targeting plasmodium falciparum invasion complex . bioRxiv , 2025 . URL https://api.semanticscholar.org/CorpusID:276407381 . 3. ↵ Rawaf Alenazy . Drug efflux pump inhibitors: A promising approach to counter multidrug resistance in gramnegative pathogens by targeting acrb protein from acrab-tolc multidrug efflux pump from escherichia coli . Biology , 11 , 2022 . URL https://api.semanticscholar.org/CorpusID:252245053 . 4. ↵ Elaheh Alipour-Khezri , Mikael Skurnik , and Gholamreza Zarrini . Pseudomonas aeruginosa bacteriophages and their clinical applications . Viruses , 16 , 2024 . URL https://api.semanticscholar.org/CorpusID:270901180 . 5. ↵ Hiroshi Arakawa , Yuya Nakazono , Natsumi Matsuoka , Momoka Hayashi , Yoshiyuki Shirasaka , Atsushi Hirao , and Ikumi Tamai . Induction of open-form bile canaliculus formation by hepatocytes for evaluation of biliary drug excretion . Communications Biology , 6 , 2023 . URL https://api.semanticscholar.org/CorpusID:261074321 . 6. Christine Årdal , Manica Balasegaram , Ramanan Laxminarayan , David McAdams , Kevin Outterson , John H. Rex , and Nithima Sumpradit . Antibiotic development — economic, regulatory and societal challenges . Nature Reviews Microbiology , 18 : 267 – 274 , 2019 . URL https://api.semanticscholar.org/CorpusID:208172222 . OpenUrl PubMed 7. Nur Ain Mohd Asri , Suhana Ahmad , Rohimah Mohamud , Nurmardhiah Mohd Hanafi , Nur Fatihah Mohd Zaidi , Ahmad Adebayo Irekeola , Rafidah Hanim Shueb , Leow Chiuan Yee , Norhayati Mohd Noor , Fatin Hamimi Mustafa , Chan Yean Yean , and Nik Yusnoraini Yusof . Global prevalence of nosocomial multidrug-resistant klebsiella pneumoniae: A systematic review and meta-analysis . Antibiotics , 10 , 2021 . URL https://api.semanticscholar.org/CorpusID:245048006 . 8. ↵ Leslie Z. Benet , Chelsea M. Hosey , Oleg Ursu , and Tudor I. Oprea . Bddcs, the rule of 5 and drugability . Advanced Drug Delivery Reviews , 101 : 89 – 98 , 2016 . ISSN 0169-409X. doi: 10.1016/j.addr.2016.05.007 . URL https://www.sciencedirect.com/science/article/pii/S0169409X16301491 . Understanding the challenges of beyond-rule-of-5 compounds. OpenUrl CrossRef PubMed 9. ↵ José Antonio Bengoechea and Joana Sá Pessoa . Klebsiella pneumoniae infection biology: living to counteract host defences . FEMS Microbiology Reviews , 43 : 123 – 144 , 2018 . URL https://api.semanticscholar.org/CorpusID:53871280 . OpenUrl 10. ↵ Vasundhra Bhandari and Akash Suresh . Next-generation approaches needed to tackle antimicrobial resistance for the development of novel therapies against the deadly pathogens . Frontiers in Pharmacology , 13 , 2022 . URL https://api.semanticscholar.org/CorpusID:249317237 . 11. G. Richard J. Bickerton , Gaia V. Paolini , Jérémy Besnard , Sorel Muresan , and Andrew L. Hopkins . Quantifying the chemical beauty of drugs . Nature chemistry , 4 2 : 90 – 8 , 2012 . URL https://api.semanticscholar.org/CorpusID:205289650 . OpenUrl CrossRef PubMed 12. ↵ Soumya Biswas , Anupam Patra , Prajita Paul , Namrata Misra , Gajraj Singh Kushwaha , and Mrutyunjay Suar . Structural and biochemical studies on klebsiella pneumoniae enoyl-acp reductase (fabi) suggest flexible substrate binding site . The Protein Journal , 43 : 84 – 95 , 2023 . URL https://api.semanticscholar.org/CorpusID:266434115 . OpenUrl PubMed 13. ↵ Olga Boiko , Martin C. Gulliford , and Caroline Burgess . Revisiting patient expectations and experiences of antibiotics in an era of antimicrobial resistance: Qualitative study . Health Expectations: An International Journal of Public Participation in Health Care and Health Policy , 23 : 1250 – 1258 , 2020 . URL https://api.semanticscholar.org/CorpusID:220531143 . OpenUrl 14. ↵ De Chang , Lokesh Sharma , Charles S. Dela Cruz , and Dong Feng Zhang . Clinical epidemiology, risk factors, and control strategies of klebsiella pneumoniae infection . Frontiers in Microbiology , 12 , 2021 . URL https://api.semanticscholar.org/CorpusID:245336338 . 15. ↵ Muyuan Chen , Xiaodong Shi , Zhili Yu , Guizhen Fan , Irina I. Serysheva , Matthew L. Baker , Ben F. Luisi , Steven J. Ludtke , and Zhao Wang . In situ structure of the acrab-tolc efflux pump at subnanometer resolution . bioRxiv , 2020 . URL https://api.semanticscholar.org/CorpusID:237479597 . 16. ↵ Antoine Daina , Olivier Michielin , and Vincent Zoete . Swissadme: a free web tool to evaluate pharmacokinetics, drug-likeness and medicinal chemistry friendliness of small molecules . Scientific Reports , 7 , 2017 . URL https://api.semanticscholar.org/CorpusID:20431782 . 17. ↵ Maxime Delmas , Magdalena Wysocka , Danilo Miranda Gusicuma , and André Freitas . Accelerating antibiotic discovery with large language models and knowledge graphs . ArXiv, abs/2503.16655 , 2025 . URL https://api.semanticscholar.org/CorpusID:277244343 . 18. ↵ Shubhangi D. Dhoble , Kashinath A. Sakhare , Pandit S. Biradar , Swati M. Narwate , Sachin P. Shinde , and Nilesh N. Shinde . From molecules to medicines: The role of artificial intelligence in accelerating drug discovery . International Journal of Advanced Research in Science, Communication and Technology , 2025 . URL https://api.semanticscholar.org/CorpusID:277729573 . 19. Jérôme Eberhardt , Diogo Santos-Martins , Andreas F. Tillack , and Stefano Forli . Autodock vina 1.2.0: New docking methods, expanded force field, and python bindings . Journal of chemical information and modeling , 2021 . URL https://api.semanticscholar.org/CorpusID:236092162 . 20. ↵ Jin Feng , Youle Zheng , Wanqing Ma , Awais Ihsan , Haihong Hao , Guyue Cheng , and Xu Wang . Multitarget antibacterial drugs: An effective strategy to combat bacterial resistance . Pharmacology & therapeutics , page 108550 , 2023 . URL https://api.semanticscholar.org/CorpusID:264810019 . 21. Christopher J L Kevin Shunji Fablina Lucien Gisela Authia Chie Murray Ikuta Sharara Swetschinski Robles Aguilar G , Christopher J. L. Murray , Kevin Shunji Ikuta , Fablina Sharara , Lucien R. Swetschinski , Gisela Robles Aguilar , Authia Gray , Chieh Han , Catherine Bisignano , Puja C. Rao , Eve E. Wool , Sarah C Johnson , Annie J. Browne , Michael Give Chipeta , Frederick Fell , Sean Hackett , Georgina Haines-Woodhouse , Bahar H. Kashef Hamadani , Emmanuelle A. P. Kumaran , Barney McManigal , Sureeruk Achalapong , Ramesh Agarwal , Samuel O Akech , Samuel B. Albertson , John Amuasi , Jason R. Andrews , A. Aravkin , Elizabeth A Ashley , F. X. Babin , Freddie Bailey , Stephen Baker , Buddha Basnyat , Adrie Bekker , Rose G. Bender , James Alexander Berkley , Adhisivam Bethou , Julia Anna Bielicki , Suppawat Boonkasidecha , James Bukosia , Cristina Gardonyi Carvalheiro , Carlos Castañeda-Orjuela , Vilada Chansamouth , Suman Chaurasia , Sara Chiurchiù , Fazle Rabbi Chowdhury , Rafai Clotaire Donatien , Aislinn Cook , Ben S Cooper , Tim Roy Cressey , Elia Criollo-Mora , Matthew Cunningham , Saffiatou Darboe , Nicholas P. J. Day , Maia De Luca , Klara Georgieva Dokova , Angela Dramowski , Susanna J. Dunachie , Thuy Duong Bich , Tim Eckmanns , Daniel Eibach , Amir Hossein Emami , Nicholas A. Feasey , Natasha Fisher-Pearson , Karen Forrest , Coralith García , Denise Garrett , Petra Gastmeier , Ababi Zergaw Giref , Rachel Claire Greer , Vikas Gupta , Sebastian Haller , Andrea Haekyung Haselbeck , Simon I. Hay , Marianne Holm , Susan Hopkins , Yingfen Hsia , Kenneth C Iregbu , Jan Jacobs , Daniel Jarovsky , Fatemeh Javanmardi , Adam Jenney , Meera Khorana , Suwimon Khusuwan , Niranjan Kissoon , Elsa Kobeissi , Tomislav Kostyanev , Fiorella Krapp , Ralf Krumkamp , Ajay Kumar , Hmwe H. Kyu , Cherry Lim , Kruy Lim , Direk Limmathurotsakul , Michael J. Loftus , Miles Lunn , Jianing Ma , Anand Manoharan , Florian Marks , Jurgen May , Mayfong Mayxay , Neema Mturi , Tatiana Munera-Huertas , Patrick Musicha , Lilian A Musila , Marisa Márcia Mussi-Pinhata , Ravi Narayan Naidu , Tomoka Nakamura , Ruchi Nimish Nanavati , Sushma Nangia , Paul N. Newton , Chanpheaktra Ngoun , Amanda Novotney , Davis Nwakanma , Christina W Obiero , Theresa J. Ochoa , Antonio Olivas-Martínez , Piero L Olliaro , Edna Ooko , Edgar Ortiz-Brizuela , Pradthana Ounchanum , Gi Deok Pak , José Luis Paredes , Anton Y. Peleg , Carlo Perrone , Thong Phe , Koukeo Phommasone , Nishad Plakkal , Alfredo Ponce de León , Mathieu Raad , Tanusha D. Ramdin , Sayaphet Rattanavong , Amy Riddell , Tamalee Roberts , Julie V. Robotham , Anna Roca , Víctor Daniel Rosenthal , Kristina E. Rudd , Neal J Russell , Helio Silva Sader , Weerawut Saengchan , Jesse Schnall , J. Anthony G. Scott , Samroeng Seekaew , Mike Sharland , Madhusudhan Demahalli Shivamallappa , José Sifuentes-Osornio , A John Simpson , Nicolas Steenkeste , Andrew James Stewardson , Temenuga Stoeva , Nidanuch Tasak , Areerat Thaiprakong , Guy E. Thwaites , Caroline Tigoi , Claudia Turner , Paul Turner , Hindrik Rogier van Doorn , Sithembiso C. Velaphi , Avina Vongpradith , Manivanh Vongsouvath , Huong X. Vu , Timothy Rutland Walsh , Judd L. Walson , Seymour Waner , Tri Wangrangsimakul , Prapass Wannapinij , Teresa M. Wozniak , Tracey E M W Young Sharma , Kalvin C. Yu , Peng Zheng , Benn Sartorius , Alan D. Lopez , Andy S. Stergachis , Catrin E. Moore , Christiane Dolecek , and Mohsen Naghavi . Global burden of bacterial antimicrobial resistance in 2019: a systematic analysis. Lancet (London , England ) , 399 : 629 – 655 , 2022 . URL https://api.semanticscholar.org/CorpusID:246077406 . OpenUrl 22. ↵ Amanda Garrido , Alban Lepailleur , Serge Mignani , Patrick Dallemagne , and Christophe Rochais. herg toxicity assessment: Useful guidelines for drug design . European journal of medicinal chemistry , 195 : 112290 , 2020 . URL https://api.semanticscholar.org/CorpusID:215758294 . 23. ↵ Sonia George , Mohammed Basheer Ramzeena , Sayee Vignesh Ram , Senthil Kumar Selvaraj , Shinu Rajan , and Thengungal Kochupappy Ravi . Design, docking, synthesis and anti e. coli screening of novel thiadiazolo thiourea derivatives as possible inhibitors of enoyl acp reductase (fabi) enzyme . Bangladesh Journal of Pharmacology , 9 : 49 – 53 , 2014 . URL https://api.semanticscholar.org/CorpusID:85285437 . OpenUrl 24. ↵ Shekina Gonzalez-Ferrer , Hernán F. Peñaloza , James A. Budnick , William G. Bain , Hayley R. Nordstrom , Janet S. Lee , and Daria Van Tyne . Finding order in the chaos: Outstanding questions in klebsiella pneumoniae pathogenesis . Infection and Immunity , 89 , 2021 . URL https://api.semanticscholar.org/CorpusID:231871980 . 25. ↵ Declan Alan Gray and Michaela Wenzel . Multitarget approaches against multiresistant superbugs . ACS Infectious Diseases , 6 : 1346 – 1365 , 2020 . URL https://api.semanticscholar.org/CorpusID:212664305 . OpenUrl PubMed 26. ↵ S Grogan and CV Preuss . Pharmacokinetics . In StatPearls [Internet] . StatPearls Publishing, Treasure Island (FL), Jul 2023 . Updated 2023 Jul 30. Available from: https://www.ncbi.nlm.nih.gov/books/NBK557744/ . 27. ↵ Emmanuel O Irek , Adewale A. Amupitan , Temitope O. Obadare , and Aaron Oladipo Aboderin . A systematic review of healthcare-associated infections in africa: An antimicrobial resistance perspective . African Journal of Laboratory Medicine , 7 , 2018 . URL https://api.semanticscholar.org/CorpusID:56481590 . 28. ↵ Soojin Jang . Acrabtolc, a major efflux pump in gram negative bacteria: toward understanding its operation mechanism . BMB Reports , 56 : 326 – 334 , 2023 . URL https://api.semanticscholar.org/CorpusID:259321871 . OpenUrl PubMed 29. ↵ Chen-Yang Jia , Jing-Yi Li , Ge-Fei Hao , and Guangfu Yang . A drug-likeness toolbox facilitates admet study in drug discovery . Drug discovery today , 2019 . URL https://api.semanticscholar.org/CorpusID:207943987 . 30. ↵ Panagiotis L. Kastritis and Alexandre M.J.J. Bonvin . On the binding affinity of macromolecular interactions: daring to ask why proteins interact . Journal of the Royal Society Interface , 10 , 2013 . URL https://api.semanticscholar.org/CorpusID:14659814 . 31. ↵ Sami Khabthani , Jean Marc Rolain , and Vicky Merhej . In silico/in vitro strategies leading to the discovery of new nonribosomal peptide and polyketide antibiotics active against human pathogens . Microorganisms , 9 , 2021 . URL https://api.semanticscholar.org/CorpusID:243830255 . 32. ↵ Chaelin Kim , Marianne Holm , Isabel Frost , Mateusz Hasso-Agopsowicz , and Kaja M. Abbas . Global and regional burden of attributable and associated bacterial antimicrobial resistance avertable by vaccination: modelling study . BMJ Global Health , 8 , 2022 . URL https://api.semanticscholar.org/CorpusID:248663270 . 33. ↵ Marta Kus , Izabela Ibragimow , and Hanna Piotrowska-Kempisty . Caco-2 cell line standardization with pharmaceutical requirements and in vitro model suitability for permeability assays . Pharmaceutics , 15 , 2023 . URL https://api.semanticscholar.org/CorpusID:264476248 . 34. ↵ Ramanan Laxminarayan , Adriano G. Dusé , Chand Wattal , Anita K. M. Zaidi , Nithima Sumpradit , Erika R. Vlieghe , Gabriel Levy Hara , Ian Malcolm Gould , Herman Goossens , Christina Greko , Maryam Bigdeli , Göran Tomson , Eva Ombaka , Arturo Quizhpe Peralta , Farah Naz Qamar , Sam Kariuki , Anthony R M Coates , and Richard J. Bergstrom . The lancet infectious diseases commission antibiotic resistance—the need for global solutions . 2013 . URL https://api.semanticscholar.org/CorpusID:212723696 . 35. ↵ Jian Li , Yuwen Shi , Xuanli Song , Xiaoyu Yin , and Hui Liu . Mechanisms of antimicrobial resistance in klebsiella: Advances in detection methods and clinical implications . Infection and Drug Resistance , 18 : 1339 – 1354 , 2025 . URL https://api.semanticscholar.org/CorpusID:276953586 . OpenUrl 36. ↵ Yanping Li , S Kumar , Lihu Zhang , and Hongjie Wu . Klebsiella pneumonia and its antibiotic resistance: A bibliometric analysis . BioMed Research International , 2022 , 2022 . URL https://api.semanticscholar.org/CorpusID:249482155 . 37. ↵ Yanping Li , S Kumar , Lihu Zhang , Hongjie Wu , and Hongyan Wu . Characteristics of antibiotic resistance mechanisms and genes of klebsiella pneumoniae . Open Medicine , 18 , 2023 . URL https://api.semanticscholar.org/CorpusID:258639645 . 38. ↵ Yanping Li , Suresh Kumar , and Lihu Zhang . Mechanisms of antibiotic resistance and developments in therapeutic strategies to combat klebsiella pneumoniae infection . Infection and Drug Resistance , 17 : 1107 – 1119 , 2024 . URL https://api.semanticscholar.org/CorpusID:268602931 . OpenUrl 39. ↵ Hao Lu and Peter J. Tonge . Inhibitors of fabi, an enzyme drug target in the bacterial fatty acid biosynthesis pathway . Accounts of chemical research , 41 1 : 11 – 20 , 2008 . URL https://api.semanticscholar.org/CorpusID:10046708 . OpenUrl CrossRef PubMed Web of Science 40. ↵ Nisha Mahey , Rushikesh Tambat , Ritu Kalia , Rajnita Ingavale , Akriti Kodesia , Nishtha Chandal , Srajan Kapoor , Dipesh Kumar Verma , Krishan Gopal Thakur , Sanjay Madhukar Jachak , and Hemraj Nandanwar . Pyrrole-based inhibitors of rnd-type efflux pumps reverse antibiotic resistance and display anti-virulence potential . PLOS Pathogens , 20 , 2024 . URL https://api.semanticscholar.org/CorpusID:269029449 . 41. ↵ Fei Mao , Wei Ni , Xiang Xu , Hui Rong Wang , Jing Wang , Min Ji , and Jian Li . Chemical structure-related drug-like criteria of global approved drugs . Molecules , 21 , 2016 . URL https://api.semanticscholar.org/CorpusID:1761024 . 42. ↵ Shahila Mehboob , Kirk E. Hevener , Kent Truong , Teuta Boci , Bernard D. Santarsiero , and Michael E. Johnson . Structural and enzymatic analyses reveal the binding mode of a novel series of francisella tularensis enoyl reductase (fabi) inhibitors . Journal of medicinal chemistry , 55 12 : 5933 – 41 , 2012 . URL https://api.semanticscholar.org/CorpusID:3156462 . OpenUrl CrossRef PubMed 43. ↵ Roberta J. Melander , Anne E. Mattingly , Ansley M. Nemeth , and Christian Melander . Overcoming intrinsic resistance in gram-negative bacteria using small molecule adjuvants . Bioorganic & medicinal chemistry letters , 80 : 129113 , 2022 . URL https://api.semanticscholar.org/CorpusID:255080263 . 44. ↵ MinohealthAILabs . Moremi ai: Towards artificial general intelligence for health (agi4health) and artificial general intelligence for biology (agi4bio) . 2025 . URL https://minohealth-storage.fra1.cdn.digitaloceanspaces.com/moremi-article.pdf . 45. PL Moja , World Health Organization , et al. Prioritization of pathogens to guide discovery, research and development of new antibiotics for drug resistant bacterial infections, including tuberculosis . 2017 . 46. ↵ Mohsen Naghavi , Stein Emil Vollset , Kevin Shunji Ikuta , Lucien R. Swetschinski , Authia Gray , Eve E. Wool , Gisela Robles Aguilar , Tomislav Metrović , Georgia Smith , Chieh Han , Rebecca L Hsu , Julian Chalek , Daniel T Araki , Erin Chung , Cat Raggi , Anna Gershberg Hayoon , Nicole Davis Weaver , Paulina A Lindstedt , Amanda E. Smith , Umut Altay , Natalia Valeryevna Bhattacharjee , Konstantinos Giannakis , Frederick Fell , Barney McManigal , Nattwut Ekapirat , Jessica Andretta Mendes , Tilleye Runghien , Oraya Srimokla , Atef Abdelkader , Sherief M. Abd-Elsalam , Richard Gyan Aboagye , Hassan Abolhassani , Hasan Abualruz , Usman Abubakar , Hana Jihad Jihad Abukhadijah , Salahdein Aburuz , Ahmed Abu-Zaid , Sureerak Achalapong , Isaac Yeboah Addo , Victor Adekanmbi , Temitayo Esther Adeyeoluwa , Qorinah Estiningtyas Sakilah Adnani , Leticia Akua Adzigbli , Muhammad Sohail Afzal , Saira Afzal , Antonella Agodi , Austin J Ahlstrom , Aqeel Ahmad , Sajjad Ahmad , Tauseef Ahmad , Alireza Ahmadi , Ayman Ahmed , Haroon Ahmed , Ibrar Ahmed , Mohammed Ahmed , Saeed Ahmed , Syed Anees Ahmed , Mohammed Ahmed Akkaif , Salah Al Awaidy , Yazan Al Thaher , Samer O. Alalalmeh , Mohammad T. Albataineh , Wafa Ali Aldhaleei , Adel Ali Saeed Al-Gheethi , Nma Bida Alhaji , Abid Ali , Liaqat Ali , Syed Shujait Shujait Ali , Waad Ali , Kasim Allel , Sabah Al-Marwani , Ahmad Alrawashdeh , Awais Altaf , Ala’a B. Al-Tammemi , Jaffar A. Al-Tawfiq , Karem H. Alzoubi , Walid Adnan Al-Zyoud , Ben Amos , John Humphrey Amuasi , Robert Viorel Ancuceanu , Jason R Andrews , Abhishek Anil , Iyadunni A. Anuoluwa , Saeid Anvari , Anayochukwu E. Anyasodor , Geminn Louis C . Apostol , Jalal Arabloo , Mosab Arafat , Aleksandr Aravkin , Demelash Areda , Abdulfatai Aremu , A. A. Artamonov , Elizabeth A. Ashley , Marvellous O Asika , Seyyed Shamsadin Athari , Maha Moh’d Wahbi Atout , Tewachew Awoke , Sina Azadnajafabad , James Mba Azam , Shahkaar Aziz , Ahmed Y. Azzam , Mahsa Babaei , François-Xavier Babin , Muhammad Badar , Atif Amin Baig , Milica Bajcetic , Stephen Baker , Mainak Bardhan , Hiba Jawdat Barqawi , Zarrin Basharat , Afisu Basiru , Mathieu Bastard , Saurav Kumar Basu , Nebiyou Simegnew Bayleyegn , Melaku Ashagrie Belete , Olorunjuwon O. Bello , Apostolos Beloukas , James A. Berkley , Akshaya Srikanth Bhagavathula , Sonu M M Bhaskar , Soumitra Sudip Bhuyan , Julia Anna Bielicki , Nikolay Ivanovich Briko , Colin Stewart Brown , Annie J. Browne , Danilo Buonsenso , Yasser Bustanji , Cristina G Carvalheiro , Carlos Castañeda-Orjuela , Muthia Cenderadewi , Joshua Chadwick , Sandip Chakraborty , Rama M Chandika , Sara Chandy , Vilada Chansamouth , Vijay Kumar Chattu , Anis Ahmad Chaudhary , Patrick R. Ching , Hitesh Chopra , Fazle Rabbi Chowdhury , Dinh Toi Chu , Muhammad Chutiyami , Natália Cruz-Martins , Alanna Gomes da Silva , Omid Dadras , Xiaochen Dai , Samuel Demissie Darcho , Saswati Das , Fernando De la Hoz , Denise Myriam Dekker , Kuldeep Dhama , Daniel Diaz , Benjamin F.R. Dickson , Serge Ghislain Djorie , Milad Dodangeh , Sushil Dohare , Klara Georgieva Dokova , Ojas Prakashbhai Doshi , Robert Kokou Dowou , Haneil Larson Dsouza , Susanna J. Dunachie , Arkadiusz Marian Dziedzic , Tim Eckmanns , Abdelaziz Ed-Dra , Aziz Eftekharimehrabad , Temitope Cyrus Ekundayo , Iman El Sayed , Muhammed Elhadi , Waseem El-Huneidi , Christelle Elias , Sally J Ellis , Randa Elsheikh , Ibrahim Elsohaby , Chadi Eltaha , Babak Eshrati , Majid Eslami , David W. Eyre , Adewale Oluwaseun Fadaka , Adeniyi Francis Fagbamigbe , Ayesha Fahim , Aliasghar Fakhri-Demeshghieh , Folorunso Oludayo Fasina , Modupe Margaret Fasina , Ali Fatehizadeh , Nicholas A. Feasey , Alireza Feizkhah , Ginenus Fekadu , Florian Fischer , Ida Fitriana , Karen M Forrest , Célia Fortuna Rodrigues , John E Fuller , Muktar Ahmed Gadanya , Márió Gajdács , Aravind P. Gandhi , Esteban Garcia-Gallo , Denise Garrett , Rupesh Kumar Gautam , Miglas Welay Gebregergis , Mesfin Gebrehiwot , Teferi Gebru Gebremeskel , Christine Geffers , Leonidas Georgalis , Ramy Mohamed Ghazy , Mahaveer Golechha , Davide Golinelli , Melita A. Gordon , Snigdha Gulati , Rajat Das Gupta , Sapna Gupta , Vijai Kumar Gupta , Awoke Derbie Habteyohannes , Sebastian Haller , Harapan Harapan , Michelle L Harrison , Ahmed I Hasaballah , Ikramul Hasan , Rumina Syeda Hasan , Hamid Hasani , Andrea Haekyung Haselbeck , Md. Saquib Hasnain , Ikrama Ibrahim Hassan , Shoaib Hassan , Mahgol Sadat Hassan Zadeh Tabatabaei , Khezar Hayat , Jiawei He , Omar E. Hegazi , Mohammad Heidari , Kamal Hezam , Ramesh Holla , Marianne Holm , Heidi Hopkins , Md Mahbub Hossain , Mehdi Hosseinzadeh , Sorin Hostiuc , Nawfal R. Hussein , Le Duc Huy , Elsa D. Ibáñez-Prada , Adalia Ikiroma , Irena M. Ilic , Sheikh Mohammed Shariful Islam , Faisal Ismail , Nahlah Elkudssiah Ismail , Chidozie Declan Iwu , Chinwe Juliana Iwu-Jaja , Abdollah Jafarzadeh , Fatoumatta Jaiteh , Reza Jalilzadeh Yengejeh , Roland Dominic G . Jamora , Javad Javidnia , Talha Jawaid , Adam W.J. Jenney , Hyon Jin Jeon , Mohammad Jokar , Nabi Jomehzadeh , Tamás Jóo , Nitin Joseph , Zul Kamal , Kehinde Kazeem Kanmodi , Rami S. Kantar , James Apollo Kapisi , Ibraheem M. Karaye , Yousef S. Khader , Himanshu Khajuria , Nauman Khalid , Faham Khamesipour , Ajmal Khan , Mohammad Jobair Khan , Muhammad Tariq Khan , Vishnu Khanal , Feriha Fatima Khidri , Jagdish Khubchandani , Suwimon Khusuwan , Min Seo Kim , Adnan Kisa , Vladimir Andreevich Korshunov , Fiorella Krapp , Ralf Krumkamp , Mohammed Kuddus , Mukhtar Kulimbet , Dewesh Kumar , Emmanuelle A P Kumaran , Ambily Kuttikkattu , Hmwe H. Kyu , Iván Landires , Basira Kankia Lawal , Thao Thi Thu Le , Ingeborg Maria Lederer , Munjae Lee , Seung Won Lee , Alain Lepape , Temesgen Leka Lerango , Virendra S. Ligade , Cherry Lim , Stephen S. Lim , Liknaw Workie Limenh , Chaojie Liu , Xiaofeng Liu , Xuefeng Liu , Michael J Loftus , Hawraz Ibrahim M Amin , Kelsey Lynn Maass , Sandeep B Maharaj , Mansour Adam Mahmoud , Panagiota Maikanti-Charalampous , Omar Mohamed Makram , Kashish Malhotra , Ahmad Azam Malik , Georgia D Mandilara , Florian Marks , Bernardo Alfonso Martínez-Guerra , Miquel Martorell , Hossein Masoumi-Asl , Alexander G. Mathioudakis , Juergen May , Theresa A McHugh , James E. Meiring , Hadush Negash Meles , Addisu Melese , Endalkachew Belayneh Melese , Giuseppe Minervini , Nouh Saad Mohamed , Shafiu Mohammed , Syam Mohan , Ali H. Mokdad , Lorenzo Monasta , AmirAli Moodi Ghalibaf , Catrin E Moore , Yousef Moradi , Elias Mossialos , Vincent Mougin , George Duke Mukoro , Francesk Mulita , Berit Muller-Pebody , Efrén Murillo-Zamora , Sani Musa , Patrick Musicha , Lillian A Musila , Saravanan Muthupandian , Ahamarshan Jayaraman Nagarajan , Pirouz Naghavi , Firzan Nainu , Tapas Sadasivan Nair , Hastyar Najmuldeen , Zuhair S. Natto , Javaid Nauman , Biswa Prakash Nayak , Gordon Takop Nchanji , Pacifique Ndishimye , Ionut Negoi , Ruxandra Irina Negoi , Seyed Aria Nejadghaderi , QuynhAnh P Nguyen , Efaq Ali Noman , Davis C. Nwakanma , Seamus O’Brien , Theresa J Ochoa , Ismail Ayoade Odetokun , Oluwaseun Adeolu Ogundijo , Tolulope R Ojo-Akosile , Sylvester Reuben Okeke , Osaretin Christabel Okonji , Andrew Toyin Olagunju , Antonio Olivas-Martínez , Abdulhakeem Olorukooba , Peter Olwoch , Kenneth Ikenna Onyedibe , Edgar Ortiz-Brizuela , Olayinka Osuolale , Pradthana Ounchanum , Oyetunde T. Oyeyemi , A Mahesh PadukudruP , José Luis Paredes , Romil R Parikh , Jay Patel , Shankargouda Patil , Shrikant Pawar , Anton Y. Peleg , Prince Peprah , João Perdigão , Carlo Perrone , Ionela-Roxana Petcu , Koukeo Phommasone , Zahra Zahid Piracha , Dimitri Poddighe , Andrew J. Pollard , Ramesh Poluru , Alfredo Ponce de León , Jagadeesh Puvvula , Farah Naz Qamar , Nameer Hashim Qasim , Clotaire Donatien Rafai , Pankaja Raghav Raghav , Leila Rahbarnia , Fakher Rahim , Vafa Rahimi-Movaghar , Mosiur Rahman , Muhammad Aziz Rahman , Hazem Ramadan , Shakthi Kumaran Ramasamy , Pushkal Sinduvadi Ramesh , Pramod Wasudev Ramteke , Rishabh Kumar Rana , Usha Rani , Mohammad Mahdi Rashidi , Devarajan Rathish , Sayaphet Rattanavong , Salman Rawaf , Elrashdy M. Redwan , Luis Felipe Reyes , Tamalee Roberts , Julie V. Robotham , Victor Daniel Rosenthal , Allen Guy Ross , Nitai Roy , Kristina E. Rudd , Cameron John Sabet , Basema Saddik , Mohammad Reza Saeb , Umar Saeed , Sahar Saeedi Moghaddam , Weeravoot Saengchan , Mohsen Safaei , Amene Saghazadeh , Narjes Saheb Sharif-Askari , Amirhossein Sahebkar , Soumya Swaroop Sahoo , Maitreyi Sahu , Morteza Saki , Nasir Salam , Zikria Saleem , Mohamed A. Saleh , Yoseph Leonardo Samodra , Abdallah M. Samy , Aswini Saravanan , Maheswar Satpathy , Austin E Schumacher , Mansour Sedighi , Samroeng Seekaew , Mahan Shafie , Pritik A. Shah , Samiah Shahid , Moyad Jamal Shahwan , Sadia Shakoor , Noga Shalev , Muhammad Aaqib Shamim , Mohammad Ali Shamshirgaran , Anas Shamsi , Amin Sharifan , Rajesh Padumane Shastry , Mahabalesh Shetty , Aminu Shittu , Sunil Shrestha , Emmanuel Edwar Siddig , Theologia Sideroglou , José Sifuentes-Osornio , Luís Manuel Lopes Rodrigues Silva , Eric A F Simões , Andrew J. H. Simpson , Amit Singh , Surjit Singh , Robert Sinto , Sameh S M Soliman , Soroush Soraneh , Nicole E Stoesser , Temenuga Stoeva , Chandan Kumar Swain , Lukasz Szarpak , Y SreeSudhaT , Shima Tabatabai , Celine Tabche , Zanan Mohammed Ameen Taha , Ker-Kan Tan , Nidanuch Tasak , Nathan Y. Tat , Areerat Thaiprakong , Pugazhenthan Thangaraju , Caroline Tigoi , Krishna Tiwari , Marcos Roberto Tovani-Palone , Thang Huu Tran , Munkhtuya Tumurkhuu , Paul Turner , Aniefiok John Udoakang , Arit Udoh , Noor Ullah , Saeed Ullah , Asokan Govindaraj Vaithinathan , Mario Valenti , Theo Vos , Huong Thi Lan Vu , Yasir Waheed , Ann Sarah Walker , Judd L. Walson , Tri Wangrangsimakul , Kosala Gayan Weerakoon , Heiman Frank Louis Wertheim , Phoebe C M Williams , Asrat Wolde , Teresa Maria Wozniak , Felicia Wu , Zenghong Wu , Mukesh Kumar Yadav , Sajad Yaghoubi , Zwanden Sule Yahaya , Amir Yarahmadi , Saber Yezli , Yazachew Engida Yismaw , Dong Keon Yon , Chun wei Yuan , Hadiza Yusuf , Fathiah Zakham , Giulia Zamagni , Haijun Zhang , Zhi-Jiang Zhang , Magdalena Zielińska , Alimuddin Zumla , Sa’ed H. Zyoud , Samer H. Zyoud , Simon I. Hay , Andy Stergachis , Benn Sartorius , Ben S Cooper , Christiane Dolecek , and Christopher J. L. Murray . Global burden of bacterial antimicrobial resistance 1990–2021: a systematic analysis with forecasts to 2050 . Lancet (London, England) , 404 : 1199 – 1226 , 2024 . URL https://api.semanticscholar.org/CorpusID:272685078 . OpenUrl CrossRef PubMed 47. ↵ Shiri Navon-Venezia , Kira Kondratyeva , and Alessandra Carattoli . Klebsiella pneumoniae: a major world-wide source and shuttle for antibiotic resistance . FEMS Microbiology Reviews , 41 : 252 – 275 , 2017 . URL https://api.semanticscholar.org/CorpusID:205077101 . OpenUrl CrossRef PubMed 48. ↵ Junio Oliveira and Wanda C. Reygaert . Gram-negative bacteria . In StatPearls Publishing, editor, StatPearls [Internet] . StatPearls Publishing, Treasure Island (FL), Aug 2023 . Updated 2023 Aug 8. Available from: https://www.ncbi.nlm.nih.gov/books/NBK538213/ . 49. ↵ JIM O’neill . Antimicrobial resistance: tackling a crisis for the health and wealth of nations . Rev. Antimicrob. Resist ., 2014 . 50. ↵ Michelle K. Paczosa and Joan Mecsas . Klebsiella pneumoniae: Going on the offense with a strong defense . Microbiology and Molecular Biology Reviews , 80 : 629 – 661 , 2016 . URL https://api.semanticscholar.org/CorpusID:7666204 . OpenUrl Abstract / FREE Full Text 51. ↵ Emma Padilla , Enrique Llobet , Antonio Doménech-Sánchez , Luis Martínez-Martínez , José Antonio Bengoechea , and Sebastián Albertí . Klebsiella pneumoniae acrab efflux pump contributes to antimicrobial resistance and virulence . Antimicrobial Agents and Chemotherapy , 54 : 177 – 183 , 2009 . URL https://api.semanticscholar.org/CorpusID:206702167 . OpenUrl PubMed 52. Joshua B Parsons and Charles O. Rock . Bacterial lipids: metabolism and membrane homeostasis . Progress in lipid research , 52 3 : 249 – 76 , 2013 . URL https://api.semanticscholar.org/CorpusID:43359463 . OpenUrl CrossRef PubMed Web of Science 53. ↵ Rainer Podschun and Uwe Ullmann . Klebsiella spp. as nosocomial pathogens: Epidemiology, taxonomy, typing methods, and pathogenicity factors . Clinical Microbiology Reviews , 11 : 589 – 603 , 1998 . URL https://api.semanticscholar.org/CorpusID:19616792 . OpenUrl Abstract / FREE Full Text 54. ↵ Preeti Rana , Shaik Mahammad Ghouse , Ravikumar Akunuri , Y.V. Madhavi , Sidharth Chopra , and Srinivas Acharya Nanduri . Fabi (enoyl acyl carrier protein reductase) – a potential broad spectrum therapeutic target and its inhibitors . European journal of medicinal chemistry , 208 : 112757 , 2020 . URL https://api.semanticscholar.org/CorpusID:221496422 . OpenUrl PubMed 55. Miah Roney and Mohd Fadhlizil Fasihi Mohd Aluwi . The importance of in-silico studies in drug discovery . Intelligent Pharmacy , 2 ( 4 ): 578 – 579 , 2024 . ISSN 2949-866X. doi: 10.1016/j.ipha.2024.01.010 . URL https://www.sciencedirect.com/science/article/pii/S2949866X24000200 . OpenUrl CrossRef 56. ↵ Md Abdus Salam , Md. Yusuf Al-Amin , Moushumi Tabassoom Salam , Jogendra Singh Pawar , Naseem Akhter , Ali A. Rabaan , and Mohammed Abdullah Alqumber . Antimicrobial resistance: A growing serious threat for global public health . Healthcare , 11 , 2023 . URL https://api.semanticscholar.org/CorpusID:259517425 . 57. Zikria Saleem , Brian B Godman , Mohamed Azmi Hassali , Furqan Khurshid Hashmi , Faiza Azhar , and Inayat Ur Rehman . Point prevalence surveys of health-care-associated infections: a systematic review . Pathogens and Global Health , 113 : 191 – 205 , 2019 . URL https://api.semanticscholar.org/CorpusID:195067027 . OpenUrl 58. ↵ Hatim F Sati , Elena Carrara , Alessia Savoldi , Paul Hansen , Jacopo Garlasco , Enrica Campagnaro , Simone Boccia , Juan Antonio Castillo-Polo , Eugenia Magrini , Pilar Garcia-Vello , Eve E. Wool , Valeria Gigante , Erin Duffy , Alessandro Cassini , Benedikt Huttner , Pilar Ramon Pardo , Mohsen Naghavi , Fuad Mirzayev , Matteo Zignol , Alexandra Cameron , and Evelina Tacconelli . The who bacterial priority pathogens list 2024: a prioritisation study to guide research, development, and public health strategies against antimicrobial resistance . The Lancet. Infectious diseases , 2025 . URL https://api.semanticscholar.org/CorpusID:277797180 . 59. ↵ Michael Schlander , Karla Hernandez-villafuerte , Chih-Yuan Cheng , Jorge Mestre-Ferrandiz , and Michael Baumann . How much does it cost to research and develop a new drug? a systematic review and assessment . Pharmacoeconomics , 39 : 1243 – 1269 , 2021 . URL https://api.semanticscholar.org/CorpusID:236953765 . OpenUrl CrossRef PubMed 60. ↵ Assar A Shah , Ameen S. S. Alwashmi , Adil Abalkhail , and Abdullah M. Alkahtani . Emerging challenges in klebsiella pneumoniae: Antimicrobial resistance and novel approach . Microbial pathogenesis , page 107399 , 2025 . URL https://api.semanticscholar.org/CorpusID:276489663 . 61. ↵ Lynn L. Silver . Challenges of antibacterial discovery . Clinical Microbiology Reviews , 24 : 71 – 109 , 2011 . URL https://api.semanticscholar.org/CorpusID:27876920 . OpenUrl Abstract / FREE Full Text 62. ↵ Kyle Swanson , Gary Liu , Denise Catacutan , Autumn Arnold , James Y. Zou , and Jonathan M Stokes . Generative ai for designing and validating easily synthesizable and structurally novel antibiotics . Nat. Mac. Intell ., 6 : 338 – 353 , 2024a . URL https://api.semanticscholar.org/CorpusID:268639779 . OpenUrl 63. ↵ Kyle Swanson , Parker Walther , Jeremy Leitz , Souhrid Mukherjee , Joseph C. Wu , Rabindra V. Shivnaraine , and James Zou . Admet-ai: a machine learning admet platform for evaluation of large-scale chemical libraries . Bioinformatics , 40 , 2024b . URL https://api.semanticscholar.org/CorpusID:270708988 . 64. Karolina Słoczyńska , Agnieszka Gunia-Krzyżak , Paulina Koczurkiewicz , Katarzyna Wójcik-Pszczoła , Dorota Żelaszczyk , Justyna Popiół , and Elżbieta Pekala . Metabolic stability and its role in the discovery of new chemical entities . Acta Pharmaceutica , 69 : 345 – 361 , 2019 . URL https://api.semanticscholar.org/CorpusID:195756676 . OpenUrl PubMed 65. ↵ Stanislav S. Terekhov , Ilya A. Osterman , and Ivan V. Smirnov . High-throughput screening of biodiversity for antibiotic discovery . Acta Naturae , 10 : 23 – 29 , 2018 . URL https://api.semanticscholar.org/CorpusID:53214457 . OpenUrl PubMed 66. Marcelo Der Torossian Torres , Yimeng Zeng , Fangping Wan , Natalie Maus , Jacob R. Gardner , and César de la Fuente-Nunez . A generative artificial intelligence approach for antibiotic optimization . bioRxiv , 2024 . URL https://api.semanticscholar.org/CorpusID:274341833 . 67. ↵ Maulikkumar D. Vaja , Heenaben A. Chokshi , Janak Jansari , Om Dixit , Shubham Savaliya , Deepak Patel , and Fenil Patel . Study of antimicrobial resistance (amr) in shigella spp. in india . Recent advances in anti-infective drug discovery , 2024 . URL https://api.semanticscholar.org/CorpusID:267466908 . 68. ↵ Jamie I. Vandenberg , Matthew D Perry , Mark Jonathan Perrin , Stefan A. Mann , Ying Ke , and Adam P . Hill. herg k(+) channels: structure, function, and clinical significance . Physiological reviews , 92 3 : 1393 – 478 , 2012 . URL https://api.semanticscholar.org/CorpusID:31857643 . OpenUrl CrossRef PubMed 69. ↵ Matthew E. Wand , Elizabeth M. Darby , Jessica M. A. Blair , and J. Mark Sutton . Contribution of the efflux pump acrab-tolc to the tolerance of chlorhexidine and other biocides in klebsiella spp . Journal of Medical Microbiology , 71 , 2022 . URL https://api.semanticscholar.org/CorpusID:247678078 . 70. ↵ Guoying Wang , Guo Zhao , Xiaoyu Chao , Longxiang Xie , and Hong ju Wang . The characteristic of virulence, biofilm and antibiotic resistance of klebsiella pneumoniae . International Journal of Environmental Research and Public Health , 17 , 2020 . URL https://api.semanticscholar.org/CorpusID:221467691 . 71. ↵ Jike Wang , Jianwen Feng , Yu Kang , Peichen Pan , Jingxuan Ge , Yan Wang , Mingyang Wang , Zhenxing Wu , Xingcai Zhang , Jiameng Yu , Xujun Zhang , Tianyue Wang , Lirong Wen , Guangning Yan , Yafeng Deng , Hui Shi , Chang-Yu Hsieh , Zhihui Jiang , and Tingjun Hou . Discovery of novel antimicrobial peptides with notable antibacterial potency by a llm-based foundation model . 2024 . URL https://api.semanticscholar.org/CorpusID:271244995 . 72. ↵ Zhao Wang , Guizhen Fan , Corey F. Hryc , James N. Blaza , Irina I. Serysheva , Michael F. Schmid , Wah Chiu , Ben F. Luisi , and Dijun Du . An allosteric transport mechanism for the acrab-tolc multidrug efflux pump . eLife , 6 , 2017 . URL https://api.semanticscholar.org/CorpusID:9915677 . 73. ↵ Reiko Watanabe , Tsuyoshi Esaki , Hitoshi Kawashima , Yayoi Natsume-Kitatani , Chioko Nagao , Rikiya Ohashi , and Kenji Mizuguchi . Predicting fraction unbound in human plasma from chemical structure: Improved accuracy in the low value ranges . Molecular pharmaceutics , 15 11 : 5302 – 5311 , 2018 . URL https://api.semanticscholar.org/CorpusID:52843702 . OpenUrl PubMed 74. ↵ Jack K Waters and Bart A. Eijkelkamp . Bacterial acquisition of host fatty acids has far-reaching implications on virulence . Microbiology and molecular biology reviews: MMBR , page e0012624 , 2024 . URL https://api.semanticscholar.org/CorpusID:273701860 . 75. ↵ World Health Organization . Antimicrobial resistance , 2023 . URL https://www.who.int/news-room/fact-sheets/detail/antimicrobial-resistance . Last accessed: 11/03/2025 . 76. ↵ Zhuang Xiang , Keyan Ding , Tianwen Lyu , Yinuo Jiang , Xiaotong Li , Zhuoyi Xiang , Zeyuan Wang , Ming Qin , Kehua Feng , Jike Wang , Qiang Zhang , and Huajun Chen . Instructbiomol: Advancing biomolecule understanding and design following human instructions . ArXiv, abs/2410.07919 , 2024 . URL https://api.semanticscholar.org/CorpusID:273233281 . 77. ↵ Min Xu , Yi qi Fu , Haishen Kong , Xiao Chen , Yu Chen , Lanjuan Li , and Qing Yang . Bloodstream infections caused by klebsiella pneumoniae: prevalence of blakpc, virulence factors and their impacts on clinical outcome . BMC Infectious Diseases , 18 , 2018 . URL https://api.semanticscholar.org/CorpusID:51889567 . 78. ↵ Jiangwei Yao , David F. Bruhn , Matthew W. Frank , Richard E. Lee , and Charles O. Rock . Activation of exogenous fatty acids to acyl-acyl carrier protein cannot bypass fabi inhibition in neisseria* . The Journal of Biological Chemistry , 291 : 171 – 181 , 2015 . URL https://api.semanticscholar.org/CorpusID:30958218 . OpenUrl PubMed 79. ↵ Hyunwoo Yoo , Bahrad A. Sokhansanj , James R. Brown , and Gail L. Rosen . Predicting anti-microbial resistance using large language models . ArXiv, abs/2401.00642 , 2024 . URL https://api.semanticscholar.org/CorpusID:266693332 . View the discussion thread. Back to top Previous Next Posted August 24, 2025. Download PDF Email Thank you for your interest in spreading the word about bioRxiv. NOTE: Your email address is requested solely to identify you as the sender of this article. Your Email * Your Name * Send To * Enter multiple addresses on separate lines or separate them with commas. You are going to email the following Moremi Bio Agent: Leveraging Agentic Large Language Model for the Discovery of Broad-Spectrum Antibiotics for Enterobacteriaceae Message Subject (Your Name) has forwarded a page to you from bioRxiv Message Body (Your Name) thought you would like to see this page from the bioRxiv website. Your Personal Message CAPTCHA This question is for testing whether or not you are a human visitor and to prevent automated spam submissions. Share Moremi Bio Agent: Leveraging Agentic Large Language Model for the Discovery of Broad-Spectrum Antibiotics for Enterobacteriaceae Gertrude Hattoh , Jeremiah Ayensu , Nyarko Prince Ofori , Solomon Eshun , Joshua Ntow Opare-Boateng , Osman Tanko , Darlington Akogo bioRxiv 2025.08.21.671656; doi: https://doi.org/10.1101/2025.08.21.671656 Share This Article: Copy Citation Tools Moremi Bio Agent: Leveraging Agentic Large Language Model for the Discovery of Broad-Spectrum Antibiotics for Enterobacteriaceae Gertrude Hattoh , Jeremiah Ayensu , Nyarko Prince Ofori , Solomon Eshun , Joshua Ntow Opare-Boateng , Osman Tanko , Darlington Akogo bioRxiv 2025.08.21.671656; doi: https://doi.org/10.1101/2025.08.21.671656 Citation Manager Formats BibTeX Bookends EasyBib EndNote (tagged) EndNote 8 (xml) Medlars Mendeley Papers RefWorks Tagged Ref Manager RIS Zotero Tweet Widget Facebook Like Google Plus One Subject Area Biochemistry Subject Areas All Articles Animal Behavior and Cognition (7620) Biochemistry (17643) Bioengineering (13866) Bioinformatics (41863) Biophysics (21411) Cancer Biology (18548) Cell Biology (25438) Clinical Trials (138) Developmental Biology (13359) Ecology (19866) Epidemiology (2067) Evolutionary Biology (24289) Genetics (15587) Genomics (22469) Immunology (17705) Microbiology (40301) Molecular Biology (17142) Neuroscience (88451) Paleontology (666) Pathology (2825) Pharmacology and Toxicology (4815) Physiology (7634) Plant Biology (15110) Scientific Communication and Education (2042) Synthetic Biology (4285) Systems Biology (9812) Zoology (2268)