Jomo Kenyatta University of Agriculture and Technology

Automatic detection of unusual behavior in videos is a challenging task. This challenge comes from its complexity and the wide range of applications it covers. Several deep learning approaches have been proposed to address this challenge. This includes recent generative methods that use deep convolutional generative adversarial networks (DCGAN). The DCGAN model has gained high research attention recently due to its performs well in extracting spatial features and solve class imbalance issue to detect abnormalities. However, a DCGAN is unstable during training and has low performance owing to its inability to capture the long‐term temporal dependency between sequences of video frames. In this study, we propose a novel gated recurrent unit (GRU)‐based DCGAN architecture to improve the training stability and performance of a DCGAN model for abnormal video behavior detection. The proposed model was trained using UCSD Ped1, UCSD Ped2, CUHK Avenue, and ShanghaiTech benchmark anomaly dataset. Compared to the DCGAN model, the proposed GRU‐based DCGAN model improved the detection accuracy and area under the curve (AUC) by an average of 19.91% and 8.57%, respectively. Compared with the 3D‐DCGAN model, the GRU‐based DCGAN model improved the detection accuracy and AUC by an average of 7.67% and 3.73%, respectively. Furthermore, the GRU‐based DCGAN model stabilized from epoch 10 and converged at epoch 38, whereas the other models remained unstable and did not converge at epoch 50. The findings highlight that the combination of GRU to enhance temporal modeling within a DCGAN framework is a logical extension to improve training stability and performance for abnormal video behavior detection.

Increase in microbial resistance has prompted evaluation of bioactive biopolymers such as chitin. This research assessed the antimicrobial and antioxidant activity of derivatives of in vitro digestion and fermentation of chitin obtained from field cricket, house cricket and black soldier fly. Chitin was chemically extracted and then in-vitro digested followed by fermentation using ABY 10 and ABT 5 probiotic cultures. Derivatives of in-vitro digestion and fermentation process were then tested for antioxidant and antimicrobial properties. The highest antioxidant activity in chitin samples fermented using ABY 10 was observed in derivatives of Gryllus bimaculatus, Acheta domesticus and Hermetia illucens chitin digested in vitro and fermented for 48 h at a concentration of 5mg/mL (61.11%, 63.88% and 61.63%). Similarly, among chitin samples fermented using ABT 5 starter culture, derivatives of H. illucens and A. domesticus chitin digested in vitro and fermented for 48 h at a concentration of 5 mg/mL exhibited the highest antioxidant activity (63.37% and 61.57%). Derivatives obtained after in vitro digestion and fermentation of the chitin samples exhibited significantly different antimicrobial activity against Escherichia coli, Vibrio cholerae, Bacteroides fragilis, Enterobacter agglomerans, Shigella dysenteriae, Staphylococcus aureus and Bacillus cereus. The antimicrobial activity of the derivatives increased with increase in fermentation time and sample concentration, with the highest activity being observed after 48 h of fermentation and at a concentration of 10 mg/mL. In conclusion, the study findings suggest that the development of chitin-based food products or the consumption of whole insects would potentially promote gut health.

Grounded in their status as a delicacy in East Africa, the trade of edible insects is increasingly acknowledged as a viable solution to food security and sustainable livelihoods. However, gender differences in trading intentions remain underexplored, particularly in informal and emerging markets. Existing research emphasizes economic and structural barriers but often overlooks the psychological and social determinants. This study addresses this gap by examining gender-specific factors influencing the intention to engage in edible insect trade, integrating demographic, socio-economic, and psychological perspectives. Using structural equation modeling (SEM), multiple regression, and an ordered probit model, we analyzed survey data from 550 traders across key food markets in Kenya, the analysis distinguished how gender moderates the effects of participation intentions. Results revealed that psychological determinants exerted a more significant influence than socio-economic variables on trading intentions. For females, perceived behavioral control (β = 0.775, p < 0.001) and descriptive norms (β = 0.536, p < 0.001) were the strongest predictors, underscoring the importance of self-efficacy and social influence. Conversely, for male traders, attitude (β = 0.331, p < 0.001) and descriptive norms (β = 0.580, p < 0.001) emerged as dominant factors, suggesting a more individualistic decision-making process. These findings highlight the necessity to enhance women’s self-efficacy through training, financing, and market access while leveraging attitudinal and social reinforcement strategies to encourage male participation. The study contributes to the behavioral economics literature on emerging markets and offers practical insights for policymakers, development agencies, and entrepreneurs seeking to promote sustainable insect-based trade.

High genomic plasticity within Escherichia coli enables it to acquire and accumulate genetic material through horizontal gene transfer. In this study, we sought to investigate the virulence genes, phylogroups, antibiotic resistance genes, plasmid replicons, multilocus sequence types (MLST), and core genome MLST of multidrug-resistant E. coli recovered from diarrheagenic children under 5 years from Mukuru Informal Settlement in Nairobi, Kenya. A total of 39 multidrug-resistant (MDR) strains had their DNA extracted, and whole-genome sequencing was done using the Illumina HiSeq 2000 platform. Twenty-six E. coli assemblies were analyzed using web-based bioinformatics tools available at the Centre for Genomic Epidemiology and EnteroBase. The isolates were categorized into four main phylogroups, where 10/26 (38.5%) belonged to the B2 phylogroup, 4/26 (15.4%) belonged to D, 3/26 (11.5%) belonged to A, 1/26 (3.8%) belonged to B1, while 8/26 (30.8%) were not determined. FimH30 was predominantly found in the most frequent phylogroup B2 and sequence type (ST) 131. The most common beta-lactam resistance genes were bla TEM-1B and bla CTXM 15 , followed by three fluoroquinolone resistance genes [ qnrS1 6/26 (23.1%), qnrB4 2/26 (7.7%), and aac(6′)-Ib-cr , 8/26 (30.8%)]. Of 26 isolates, 15 had at least one amino acid substitution in the housekeeping genes gyrA ( p.S83L ), gyrA ( p.D87N ), parC ( p.S80I ), parC ( p.E84V ), parC ( p.S57T ), and parE ( p.I529L ), associated with resistance to fluoroquinolones. A total of 40 diverse virulence genes were detected among the isolates. Thirteen different STs were isolated from the E. coli genomes, which included ST 131, ST 3036, ST 38, ST 10, ST 12569, ST 15271, ST 2076, ST 311, ST 3572, ST 394, ST 453, ST 46, and ST 1722. Only two isolates (2/26, 7.7%) from the Municipal City Council clinic were genetically related. Additionally, the most abundant plasmid replicon identified belonged to the IncF family, IncFII(pRSB107), in particular, followed by the Col family. The study highlighted the first E. coli ST46 to harbor the bla NDM5 gene encoded in Col(BS512), IncFII(pRSB107), and IncFIB(AP001918) plasmid replicons in Kenya. We further demonstrated the diversity of MDR E. coli associated with diarrhea in an endemic setting in Kenya. IMPORTANCE This study investigated the molecular characterization of multidrug-resistant Escherichia coli isolated from diarrheagenic children under 5 years of age in Mukuru Informal Settlement in Nairobi, Kenya. This is an important addition to the genomic analysis data of multi-drug resistant diarrheal Escherichia coli in Kenya. The use of whole-genome sequencing to identify and characterize these isolates is valuable and provides valuable insights into the molecular epidemiology of E. coli in the region.

Tuberculosis (TB) is one of the infectious diseases of public health concern globally. Kenya is ranked 15th among the 22 high TB burden countries worldwide, which collectively contribute to 80% of the world’s TB cases. TB Treatment failure is one of the threats to the control of TB. The research aimed at determining affordable predictors of TB treatment failure in a resource limited setting to inform policy in designing public health interventions that are best suited to the country’s needs. To determine the predictors of treatment failure among patients with sputum smear positive pulmonary TB attending selected public health facilities in Nairobi Count. Data was abstracted and summarized from both patients and their medical records, focusing on socio-demographic, behavioral, and clinical exposure data. Data was collected from 4 Sub-counties, a total of 21 public health facilities with high case load of pulmonary TB were reached. Utilizing an unmatched case-control design, the study enrolled 81 patients diagnosed with TB treatment failure (cases) and 162 patients who were declared cured after completing their anti-TB treatment (controls. Strengthen contact tracing, screening, and documentation of TB treatment failure cases. Conduct further studies to elucidate the association between HIV and TB treatment failure. The factors significantly associated with treatment failure in this study encompassed prior exposure to first-line anti-Tuberculosis drugs, positive sputum smear at 2 months of treatment, and suboptimal adherence to anti-TB treatment. These findings contribute valuable insights into the identification of simple predictors of TB treatment failure such as utilizing sputum microscopy or gene expert testing at 2 months of treatment to detect individuals at risk and strengthen the implementation of DOT and TB treatment failure contact tracing protocol.

Background The coronavirus disease 2019 (COVID-19) pandemic underscored the global need for reliable diagnostic tools with quick turnaround time for effective patient management and mitigation of virus spread. This study aimed to express severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) nucleocapsid protein and produce monoclonal antibodies (mAbs) against the expressed protein. Methods Following successful expression and purification of His-tagged SARS-CoV-2 N protein using a wheat germ cell-free protein expression system (WGCFS), BALB/c mice were immunized, and generated hybridomas screened for mAb production. Indirect and sandwich ELISA were used to screen the reactivity of the monoclonal antibody against both our recombinant antigen and commercial antigen. The mAbs were also assessed for their performance using RT-PCR confirmed positive samples with varying cycle threshold (CT) values and their specificity screened using virus isolates of other respiratory viruses. Results Our mAb demonstrated high reactivity against our recombinant antigen, commercial antigen, SARS-CoV-2 Beta and Omicron variants. There was no significant difference in the binding affinity of our mAb and commercial mAb against the study recombinant (p = 0.12) and commercial (p = 0.072) antigens. Our mAb detected SARS-CoV-2 from clinical samples with varying CT values and exhibited no cross-reactivity against other respiratory viruses. Conclusions We successfully expressed SARS-CoV-2 N protein leveraging WGCFS in a resource-limited setting. Our mAb had a high binding affinity to the recombinant antigen, making it a suitable candidate for antigen detection kit development. Beyond diagnostics, the mAb holds potential for therapeutic applications as well as use in clinical and environmental surveillance platforms.

The mango processing industry generates peels as byproducts, which are a rich source of bioactive compounds, including pectin. This study aimed to evaluate the impact of extraction methods on antioxidants and pectin yields from mango peels to determine the most effective techniques for recovery of bioactive compounds for potential applications in food industry. The antioxidants were extracted using microwave assisted extraction (MAE), enzyme assisted extraction (EAE), ultrasound assisted extraction (UAE) and water-bath assisted extraction (WBAE) while pectin extraction was performed using MAE, conventional acid extraction (CAE) and UAE. Drying methods (solar, oven, freeze) and mango varieties (Apple, Ngoe) were also analyzed for their influence on pectin yield. Antioxidant analysis revealed similar total phenolic content (TPC) across MAE, UAE, and WBAE, with EAE and UAE showing the highest DPPH scavenging activity (84%). Pectin yield was highest for the CAE method (21.5%; optimal conditions: pH 1.5, 80 °C, 1 h), followed by MAE (19%; pH 1.5, 900 W, 6 min) and UAE (8%; pH 1.5, 40 kHz, 30 min). The degree of esterification (DE) of MAE pectin (54%) was significantly higher than CAE extracted pectin (47%). MAE pectin exhibited a water-holding capacity (WHC) of 7.29 g water/g sample, comparable to 8.05 g for CAE pectin. Significant differences were observed in oil-holding capacity (OHC) and swelling capacity (SWC), with MAE pectin showing higher soluble dietary fiber content and swelling ability. Freeze-drying and the Apple variety gave slightly higher pectin yields than other drying methods and the Ngoe variety, respectively, though the differences were not significant. Fourier-transform infrared (FTIR) spectroscopy confirmed the structural integrity of the extracted pectin, with characteristic hydroxyl (-OH), carbonyl (C = O), and glycosidic (C–O–C) absorption bands. The absence of lipid or wax contamination in all pectin samples confirmed their purity. This study highlights the potential of mango peel pectin in waste valorization aimed for food applications.

Potato (Solanum tuberosum L.) is an important food crop in Kenya, providing a source of nutrition and income for many farmers. However, potato cyst nematodes (PCN) cause significant damage to potato plants, leading to substantial economic losses and threatening the nation’s food security. Understanding the composition and functional potential of bacterial communities in the soil is important for developing sustainable biological control strategies against PCN and improving soil health. This cross-sectional purposive study examined the bacterial communities associated with PCN-suppressive and conducive potato rhizosphere from two major potato-producing counties in Kenya. We analyzed 180 soil samples from symptomatic and asymptomatic potato plants using shotgun metagenomics, followed by functional analysis to identify genes and metabolic pathways relevant to soil and plant health. Taxonomic classification revealed Enterobacteriaceae and Pseudomonadaceae as the most dominant bacterial families present. Within these families, the genera Pseudomonas and Enterobacter were highly abundant, both known for their plant growth-promoting traits, including biological control of soil pathogens and nutrient solubilization. KEGG and Pfam database analysis revealed pathways associated with nutrient cycling, transport systems, and metabolic functions. The abundance of iron-acquisition, chemotaxis, and diverse transport genes across analyzed samples suggests the presence of beneficial bacterial communities. This study provides the first report on bacterial ecology in PCN-infested rhizosphere in Kenya and its implications for soil health and PCN management.

BACKGROUND AND OBJECTIVES Work-life balance (WLB) is the individual's view that personal and professional activities in their life align with current life priorities. WLB is important for health and is thought to prevent burnout in the workplace. Although high rates of burnout exist in neurosurgery (NS), studies of WLB and the factors that influence WLB in NS are not known. METHODS An electronic international survey using a physician wellness framework was conducted globally. χ ² tests were used to analyze the association between WLB and age, sex, level of practice, and continent of practice. RESULTS Of 446 respondents (65% staff, 35% trainees; median age range 35-44 years age category; 28% women), only 42% indicated the presence WLB. The presence of WLB was significantly lower in trainees compared with staff (χ ² = 14.065, P = .0002, odds ratio [OR]: 0.45 [95% CI: 0.30-0.68]), those aged 44 years and below (χ ² = 4.1464, P = .04172; OR: 0.63 [95% CI: 0.41-0.96]), and those in the African region compared with non-African region (χ ² = 8.33, P = .0039, OR: 0.42 [95% CI: 0.24-0.75]). CONCLUSION Nearly two-thirds of those in NS report poor WLB with trainees and younger individuals at particular risk. Lack of sufficient numbers of neurosurgeons for the workload and the lack of support staff require urgent attention globally. There is an urgent need for healthcare organizations globally to take leadership in implementing practices to improve WLB. Evidence shows these changes will likely improve personal and organizational well-being, retention, and improve medical student interest in NS.

The antimicrobial resistance (AMR) crisis represents a significant global threat. Unlike traditional antibiotics, antimicrobial peptides offer a promising pathway because of their primary mechanisms. This study aimed to evaluate and rationally design novel AMPs based on tobacco nectar's AMP (Pep 6) to combat antibiotic resistance issues. Substitution and truncation of some amino acids were applied. Four peptides, KF19, KF16, LK16, and LR16, were designed with enhanced net charge hydrophobicity. They were evaluated for their in vitro antibacterial activity. However, only promising AMPs were further evaluated for their hemolytic activity, time-killing kinetics, mode of action, and anti-biofilm properties. The results showed that only KF19 and LR16 have potent activity against Staphylococcus aureus ATCC25923 and resistant isolates with MIC values from 7.81 to 15.62 μg/mL. Hemolysis ratios were 2.38% and 2.24% at 125 μg/mL for KF19 and LR16, respectively. Both peptides were able to kill S. aureus ATCC25923 within 2 h. SEM results showed their ability to target the cell membrane. Both peptides destroyed the S. aureus biofilms significantly at 62.5 and 125 μg/mL (**p < 0.01, ***p < 0.001, ****p < 0.0001). This study supported rational design in developing new antibacterial agents and demonstrated the therapeutic potency of novel peptides that could solve the resistance issues.

The characterization of Lamu offshore reservoirs remains limited due to the absence of integrated studies combining petrophysical analysis and rock physics modeling. This study aims to enhance reservoir characterization, reduce exploration risks, and provide a framework for similar geological settings. Log data from three wells were analyzed to determine key petrophysical properties and evaluate rock physics models for lithology and fluid discrimination. Reservoir zones were delineated based on petrophysical parameters, including clay volume, porosity, hydrocarbon saturation, and gamma ray and resistivity responses. The selected reservoirs exhibited favorable characteristics, with low shale volume (0.07–0.26), high effective porosity (0.12–0.25), low water saturation (0.23–0.56), and a net thickness (18.95–43.22 m). Rock physics cross-plots (mu-rho vs. density, acoustic impedance vs. lambda-rho, and Vp/Vs ratio vs. acoustic impedance, among others) effectively distinguished hydrocarbon-bearing zones from brine-saturated sands and shales. Color-coded cross-plots further validated fluid discrimination, showing low water saturation and gamma ray values with high porosity in hydrocarbon zones. Gassmann fluid substitution analysis confirmed that replacing water with hydrocarbons significantly reduced density and had a more pronounced effect on compressional velocity than shear velocity. These findings highlight an integrated approach to minimizing hydrocarbon exploration risks, particularly in avoiding dry wells, and offer valuable insights for future exploration efforts in Lamu offshore and similar basins.

Incorporation of a smart antenna system (SAS) in wireless communication presents an array of benefits such as the elimination of interference, good security and enhanced utilisation of frequency spectrum. It is, however, important to note that these benefits can be boosted further with the application of optimisation algorithms. In this study, the beam pattern of the uniform rectangular array (URA) geometry is optimised using a proposed hybrid algorithm, which consists of the particle swarm optimisation (PSO) algorithm and the honey badger algorithm (HBA). MATLAB was utilised in the simulative analysis and optimisation process. The developed system performance is evaluated in two stages: First, the analysis of the nonoptimised URA beam pattern is conducted, followed by the analysis of the optimised beam pattern using the proposed hybrid algorithm. The hybrid algorithm performance is compared with HBA, PSO, Gaussian quantum‐behaved PSO (GQPSO), differential evolution algorithm (DEA) and the nondominated sorting genetic algorithm III (NSGA‐III). The cost function is used 50 times, and the best results of each optimisation algorithm are presented. A laptop with an 11th Gen Intel(R) Core (TM) i7‐1165G7 @ 2.80 GHz processor was used for the analysis. Results of 50 × 50 elements and 0.5λ interelement spacing show that the hybrid algorithm performs better than the other algorithms across various metrics. Specifically, it minimises the side lobe levels (SLL) from −24.29 dB (nonoptimised) to −31.22 dB, surpassing GQPSO (−26.41 dB), PSO (−29.23 dB), HBA (−31.14 dB), DEA (−28.51 dB) and the NSGA‐III (−30.88 dB). The signal‐to‐noise ratio (SNR) is increased from 24.28 dB (nonoptimised) to 31.20 dB, outperforming GQPSO (26.39 dB), PSO (29.21 dB), HBA (31.12 dB), DEA (28.50 dB) and the NSGA‐III (30.86 dB). Regarding computation time, the hybrid algorithm needs 1.30 s, which is competitive in comparison to the other algorithms.

Background The aim of this study was to assess the population and health system factors affecting the transferability of health kiosks in markets in Kenya. Methods A cross-sectional study with a partially mixed concurrent dominant status design was conducted among 843 households, policy actors, market chairpersons and champions, community health promoters and health workers. A χ2 test was used to test for independence with variables with a statistical significance (p<0.05) subjected to logistic regression. Qualitative data were transcribed verbatim to form nodes and themes. Results Level of income, knowledge, awareness and perception of cardiovascular disease (CVD) risk factors were associated with 27.5% of Nyeri respondents earning ≥Ksh 10 000 monthly compared with Vihiga respondents (17.3%). Vihiga respondents were likely to identify excessive alcohol consumption as a cause of CVD. Vihiga had fair (40%) and good (26.6%) awareness levels towards CVDs compared with Nyeri respondents (36.3% and 19.7%, respectively). Vihiga respondents had a higher positive perception towards health services at the local facility compared with Nyeri respondents. Conclusions CVD burden, low awareness levels, low health insurance cover and the poor attitude of health workers have the potential to affect the transferability of a health intervention such as a health market kiosk.

The Mount Kenya forest ecosystem (MKFE), a crucial biodiversity hotspot and one of Kenya’s key water towers, is increasingly threatened by climate change, putting its ecological integrity and vital ecosystem services at risk. Understanding the interactions between climate extremes and forest dynamics is essential for conservation planning, especially in the Mount Kenya Forest Ecosystem (MKFE), where rising temperatures and erratic rainfall are altering vegetation patterns, reducing forest resilience, and threatening both biodiversity and water security. This study integrates remote sensing and machine learning to assess historical vegetation changes and predict areas at risk in the future. Landsat imagery from 2000 to 2020 was used to derive vegetation indices comprising the Normalized Difference Vegetation Index (NDVI), Enhanced Vegetation Index (EVI), Soil-Adjusted Vegetation Index (SAVI), and Bare Soil Index (BSI). Climate variables, including extreme precipitation and temperature indices, were extracted from CHIRPS and ERA5 datasets. Machine learning models, including Random Forest (RF), XGBoost, and Support Vector Machines (SVM), were trained to assess climate-vegetation relationships and predict future vegetation dynamics under the SSP245 climate scenario using Coupled Model Intercomparison Project Phase 6 (CMIP6) downscaled projections. The RF model achieved high accuracy (R² = 0.82, RMSE = 0.15) in predicting the dynamics of vegetation conditions. Model projections show a 49–55% decline in EVI across forest areas by 2040, with the most pronounced losses likely in lower montane zones, which are more sensitive to climate-induced vegetation stress. Results emphasize the critical role of precipitation in sustaining forest health and highlight the urgent need for adaptive management strategies, including afforestation, sustainable land-use planning, and policy-driven conservation efforts. This study provides a scalable framework for modelling climate impacts on forest ecosystems globally and offers actionable insights for policymakers.

Stomoxys flies are widely distributed and economically significant vectors of various livestock pathogens of veterinary importance. However, the role of Stomoxys spp. in pathogen transmission is poorly understood. Therefore, we studied the feeding patterns of these blood feeders collected from specific locations in Kenya to identify various vertebrate hosts they fed on, the livestock hemopathogens they carried, and to elucidate their role in pathogens transmission. Our findings show that field-collected Stomoxys flies carried several pathogens, including Trypanosoma spp., Anaplasma spp., and Theileria spp., which were also detected in the blood of sampled livestock, namely camels and cattle. The findings on blood meal analysis show that Stomoxys flies fed on various domestic and wild vertebrate hosts. We further determined whether Stomoxys spp. are vectors of hemopathogens they harbored by studying the vector competence of Stomoxys calcitrans, S. niger niger, and S. boueti species complex, through laboratory and natural experimental in vivo studies. We show that in the process of blood feeding Stomoxys spp. complexes can transmit Trypanosoma evansi (8.3%) and T. vivax (30%) to Swiss white mice. In addition, field-collected Stomoxys spp. were exposed to healthy mice for blood meal acquisition, and in the process of feeding, they transmitted Theileria mutans and Anaplasma spp. to Swiss white mice (100% infection in the test mice group). All mice infected with trypanosomes via Stomoxys bite died while those infected with Theileria and Anaplasma species did not, demonstrating the virulence difference between pathogens. The key finding of this study showing the wide distribution, broad feeding host range, plethora of pathogens harbored, and efficient vector competence in spreading multiple pathogens suggests the significant role of Stomoxys on pathogen transmission and infection prevalence in livestock.

In this paper, the mathematical theory of elasticity that enables the construction of representative stress functions for photoelastic experimental hybrid method (PEHM) is revisited and reviewed. PEHM has been shown as an important and powerful tool used by experimental stress analysts to predict the stress state in complex engineering structures. To demonstrate the utility of stress functions from the mathematical theory of elasticity in real engineering applications, the contact problem of a mechanical seal with a rectangular cross-section as well as a plate with a central hole are considered. It was found that when the stress functions are applied to the contact problem of a mechanical seal with rectangular cross section, the contact stresses on the upper side were larger compared to those on the front side. The highest stresses on the front side were concentrated in the region around the extrusion gap. When a comparison between theoretical and experimental stress concentration factors (SCF) was done, it was found that there was remarkable agreement between theoretical and experimental results. Therefore, the mathematical theory of elasticity from this study shows that it can provide stress functions that serve as an invaluable input tool to predict the SCF using the photoelastic experimental hybrid method.

Worldwide, drought significantly impacts crop output and agricultural productivity. To mitigate this, eco-efficient methods to enhance plant growth under drought stress are essential. This study isolated 30 morphologically distinct bacterial strains from the maize rhizosphere in drought-stressed conditions in Makueni County, Kenya. These strains’ drought tolerance was assessed using three stress treatments of polyethylene glycol (PEG 6000) at 0, −0.75, and −1.3 MPa. The two best isolates, MK6 (Bacillus cereus) and MK17 (Bacillus velezensis), were identified via 16S rRNA gene sequencing, with sequences deposited in the NCBI database (accession numbers PP064566 and PP064918, respectively). Isolates were screened for plant growth promotion (PGP) properties under both stress and no-stress conditions, including indole acetic acid, gibberellic acid, exopolysaccharide, siderophore, antioxidant enzyme activity, proline, salicylic acid, biofilm, and biosurfactant production. Results under PEG 6000–induced stress showed MK6 produced the highest GA (51.67 µgmL⁻¹) and IAA (2.6 µgmL⁻¹). MK6 also had the highest EPS production (4320 µgmL⁻¹) and siderophore production (68%) at −1.3 MPa. Antioxidant activities, notably catalase, were highest in MK17 (1260 µgmL⁻¹). The emulsification index (%EI 24) for MK6 and MK17 was 16.72% and 10.36% under no-stress conditions and declined during stress. The combinatorial potential of MK6 and MK17 showed significant improvement in maize vigor index, germination rate, seedling length, height, and dry biomass. Therefore, MK6 and MK17 hold great potential for developing biostimulants to ameliorate drought stress in drought-prone regions of the world and promoting sustainable crop production.

Global climate change is projected to disproportionately impact cereal crop yields in developing regions, such as Kenya, due to increased vulnerability and limited adaptation capacity of the population. This study examines the current and projected influence of climate change on maize yields in two major maize-producing counties of Kenya. Utilizing the calibrated and evaluated DSSAT-CERES-Maize model (where DSSAT is Decision Support Systems for Agrotechnology Transfer and CERES stands for Crop and Environment REsource Synthesis) for the H614 maize cultivar, we investigated the projected impact of climate change on maize production with reference to a baseline period (1984–2013). Simulations were conducted for the mid-century period (2041–2070) and end-of-century period (2071–2100) using projected climate data from regional climate models (RCMs) under two Representative Concentration Pathways (RCPs; 4.5 and 8.5) scenarios. Our findings indicate a substantial decline in maize yields, ranging from 7 to 20% for the mid-century period and between 22 and 41% for the end-of-century period, with increased temperature during critical growth phases identified as the primary driver. Spatial clustering and hotspot analysis reveal differential climate impacts across the region. In the end-of-century period, both scenarios revealed that the counties will be marked by hotspots and adaptation spots, areas where climate change adaptation should be intensified. The study underscores the urgency for tailored, location-specific adaptation measures such as maize-legume intercropping, drought-resistant crops, soil water conservation and optimum sowing to mitigate future yield losses and adapt maize production to climate change.