Growth Stage-Dependent Variation in Soil Quality and Microbial Diversity of Ancient Gleditsia sinensis.

The environment monitoring of forest is vital for the ecosystem sustainable management, especially soil quality. Ancient Gleditsia sinensis is one of the most distributed ancient trees in Shaanxi. Comprehensive soil evaluate is important for the ancient tree protection. In this study, we selected the most distributed ancient tree Gleditsia sinensis and immature tree to compare the effect of growth stage to soil quality and soil bacteria. Most ancient tree soil nutrients were in good condition compared with immature tree. The bacterial community were composed with Proteobacteria (27.55%), Acidobacteriota (16.82%), Actinobacteriota (15.77%), Gemmatimonadota (6.82%), Crenarchaeota (4.61%), Bacteroidota (4.41%), Firmicutes (4.32%), Chloroflexi (4.28%), Planctomycetota (3.24%) and Verrucomicrobiota (3.04%). The level 2 ancient tree (300-400 years old) was different in bacterial community diversity. SOC and STN were important to level 2 (300-400 years old Gleditsia sinensis), and other levels were opposite. Our results suggested that the ancient tree management should not be lumped together.

A whole process study of dual microalgae cultivation coupled to domestic wastewater treatment and wheat growth.

The multiple microalgal collaborative treatment of domestic wastewater has been extensively investigated, but its whole life cycle tracking and consequent potential have not been fully explored. Herein, a dual microalgal system was employed for domestic wastewater treatment, tracking the variation in microalgal growth and pollutants removal from shake flask scale to 18 L photobioreactors scales. The results showed that Chlorella sp. HL and Scenedesmus sp. LX1 combination had superior growth and water purification performance, and the interspecies soluble algal products promoted their growth. Through microalgae mixing ratio and inoculum size optimized, the highest biomass yield (0.42 ± 0.03 g/L) and over 91 % N, P removal rates were achieved in 18 L photobioreactor. Harvested microalgae treated in different forms all promoted wheat growth and suppressed yellow leaf rate. This study provided data support for the whole process tracking of dual microalgal system in treating domestic wastewater and improving wheat growth.

Pyrroloquinoline quinone production defines the ability of Devosia species to degrade deoxynivalenol.

Deoxynivalenol (DON) is a prevalent mycotoxin that primarily contaminates cereal crops and animal feed, posing a significant risk to human and animal health. In recent years, an increasing number of Devosia strains have been identified as DON degradation bacteria, and significant efforts have been made to explore their potential applications in the food and animal feed industries. However, the characteristics and mechanisms of DON degradation in Devosia strains are still unclear. In this study, we identified a novel DON degrading bacterium, Devosia sp. D-G15 (D-G15), from soil samples. The major degradation products of DON in D-G15 were 3-keto-DON, 3-epi-DON and an unidentified product, compound C. The cell viability assay showed that the DON degradation product of D-G15 revealed significantly reduced toxicity to HEK293T cells compared to DON. Three enzymes for DON degradation were further identified, with G15-DDH converting DON to 3-keto-DON and G15-AKR1/G15-AKR6 reducing 3-keto-DON to 3-epi-DON. Interestingly, genome comparison of Devosia strains showed that the pyrroloquinoline quinone (PQQ) synthesis gene cluster is a unique feature of DON degradation strains. Subsequently, adding PQQ to the cultural media of Devosia strains without PQQ synthesis genes endowed them with DON degradation activity. Furthermore, a novel DON-degrading enzyme G13-DDH (<30% homology with known DON dehydrogenase) was identified from a Devosia strain that lacks PQQ synthesis ability. In summary, a novel DON degrading Devosia strain and its key enzymes were identified, and PQQ production was found as a distinct feature among Devosia strains with DON degradation activity, which is important for developing Devosia strain-based technology in DON detoxification.

Ageing microenvironment mediates lymphocyte carcinogenesis and lymphoma drug resistance: From mechanisms to clinical therapy (Review).

Cellular senescence has a complex role in lymphocyte carcinogenesis and drug resistance of lymphomas. Senescent lymphoma cells combine with immunocytes to create an ageing environment that can be reprogrammed with a senescence­associated secretory phenotype, which gradually promotes therapeutic resistance. Certain signalling pathways, such as the NF­κB, Wnt and PI3K/AKT/mTOR pathways, regulate the tumour ageing microenvironment and induce the proliferation and progression of lymphoma cells. Therefore, targeting senescence­related enzymes or their signal transduction pathways may overcome radiotherapy or chemotherapy resistance and enhance the efficacy of relapsed/refractory lymphoma treatments. Mechanisms underlying drug resistance in lymphomas are complex. The ageing microenvironment is a novel factor that contributes to drug resistance in lymphomas. In terms of clinical translation, some senolytics have been used in clinical trials on patients with relapsed or refractory lymphoma. Combining immunotherapy with epigenetic drugs may achieve better therapeutic effects; however, senescent cells exhibit considerable heterogeneity and lymphoma has several subtypes. Extensive research is necessary to achieve the practical application of senolytics in relapsed or refractory lymphomas. This review summarises the mechanisms of senescence­associated drug resistance in lymphoma, as well as emerging strategies using senolytics, to overcome therapeutic resistance in lymphoma.

High capacities of carbon capture and photosynthesis of a novel organic carbon-fixing microalgae in municipal wastewater: From mutagenesis, screening, ability evaluation to mechanism analysis.

The development of wastewater treatment processes capable of reducing and fixing carbon is currently a hot topic in the wastewater treatment field. Microalgae possess a natural carbon-fixing advantage, and microalgae that can symbiotically coexist with indigenous bacteria in actual wastewater attract more significant attention. Ultraviolet (UV) mutagenesis and dissolved organic carbon (DOC) acclimation were applied to strengthen the carbon-fixing performance of microalgae in this study. The mechanisms associated with microalgal water purification ability, gene regulation at the molecular level and photosynthetic potential under different trophic modes resulting from carbon fixation and transformation were disclosed. The superior performance of Chlorella sp. MHQ2 was eventually screened out among a large number of mutants generated from 3 wild-type Chlorella strains. Results indicated that the dry cell weight of the optimal species Chlorella sp. HQ mutant MHQ2 was 1.91 times that of the wild strain in the pure algal system, more carbon from municipal wastewater (MW) were transferred to the microalgae and re-entered into the biological cycle through resource utilization. In addition, COD, NH3-N and TP removal efficiencies of MW by Chlorella sp. MHQ2 were found to increase to 95.8% (1.1-times), 96.4% (1.4-times), and 92.9% (1.2-times), respectively, under the extra DOC supply and the assistance of indigenous bacteria in the MW. In the transcriptome analysis of the logarithmic phase, the glycolytic pathway was inhibited, and the pentose phosphate pathway was mainly carried out for microalgal life activities, further promoting efficient energy utilization. Upon analysis of carbon capture capacity and photosynthetic potential in trophic mode, the addition of NaHCO3 increased the photosynthetic rate of Chlorella sp. MHQ2 in mixotrophy whereas it was attenuated in autotrophy. This study could provide a new perspective for the study of resource utilization and microalgae carbon- fixing mechanisms in the actual wastewater treatment process.

Transrectal versus transperineal prostate biopsy for cancer detection in patients with gray-zone prostate-specific antigen: a multicenter, real-world study.

Knowledge about the effect of different prostate biopsy approaches on the prostate cancer detection rate (CDR) in patients with gray-zone prostate-specific antigen (PSA) is limited. We performed this study to compare the CDR among patients who underwent different biopsy approaches and had rising PSA levels in the gray zone. Two hundred and twenty-two patients who underwent transrectal prostate biopsy (TRB) and 216 patients who underwent transperineal prostate biopsy (TPB) between June 2016 and September 2022 were reviewed in this study. In addition, 110 patients who received additional targeted biopsies following the systematic TPB were identified. Clinical parameters, including age, PSA derivative, prostate volume (PV), and needle core count, were recorded. The data were fitted via propensity score matching (PSM), adjusting for potential confounders. TPB outperformed TRB in terms of the CDR (49.6% vs 28.3%, P = 0.001). The clinically significant prostate cancer (csPCa) detection rate was not significantly different between TPB and TRB (78.6% vs 68.8%, P = 0.306). In stratified analysis, TPB outperformed TRB in CDR when the age of patients was 65-75 years (59.0% vs 22.0%, P < 0.001), when PV was 25.00-50.00 ml (63.2% vs 28.3%, P < 0.001), and when needle core count was no more than 12 (58.5% vs 31.5%, P = 0.005). The CDR ( P = 0.712) and detection rate of csPCa ( P = 0.993) did not significantly differ among the systematic, targeted, and combined biopsies. TPB outperformed TRB in CDR for patients with gray-zone PSA. Moreover, performing target biopsy after systematic TPB provided no additional benefits in CDR.

Enhancing missense variant pathogenicity prediction with protein language models using VariPred.

Computational approaches for predicting the pathogenicity of genetic variants have advanced in recent years. These methods enable researchers to determine the possible clinical impact of rare and novel variants. Historically these prediction methods used hand-crafted features based on structural, evolutionary, or physiochemical properties of the variant. In this study we propose a novel framework that leverages the power of pre-trained protein language models to predict variant pathogenicity. We show that our approach VariPred (Variant impact Predictor) outperforms current state-of-the-art methods by using an end-to-end model that only requires the protein sequence as input. Using one of the best-performing protein language models (ESM-1b), we establish a robust classifier that requires no calculation of structural features or multiple sequence alignments. We compare the performance of VariPred with other representative models including 3Cnet, Polyphen-2, REVEL, MetaLR, FATHMM and ESM variant. VariPred performs as well as, or in most cases better than these other predictors using six variant impact prediction benchmarks despite requiring only sequence data and no pre-processing of the data.

Improved hypertensive stroke classification based on multi-scale feature fusion of head axial CT angiogram and multimodal learning.

PURPOSE: Strokes are severe cardiovascular and circulatory diseases with two main types: ischemic and hemorrhagic. Clinically, brain images such as computed tomography (CT) and computed tomography angiography (CTA) are widely used to recognize stroke types. However, few studies have combined imaging and clinical data to classify stroke or consider a factor as an Independent etiology. METHODS: In this work, we propose a classification model that automatically distinguishes stroke types with hypertension as an independent etiology based on brain imaging and clinical data. We first present a preprocessing workflow for head axial CT angiograms, including noise reduction and feature enhancement of the images, followed by an extraction of regions of interest. Next, we develop a multi-scale feature fusion model that combines the location information of position features and the semantic information of deep features. Furthermore, we integrate brain imaging with clinical information through a multimodal learning model to achieve more reliable results. RESULTS: Experimental results show our proposed models outperform state-of-the-art models on real imaging and clinical data, which reveals the potential of multimodal learning in brain disease diagnosis. CONCLUSION: The proposed methodologies can be extended to create AI-driven diagnostic assistance technology for categorizing strokes.

Real-world analysis of apalutamide-associated skin rash in Chinese patients with prostate cancer.

PURPOSE: This study aimed to explore the clinical characteristics of apalutamide-associated skin rash and management of skin rash in real-world Chinese patients with prostate cancer. METHODS: We investigated 138 patients with prostate cancer who received apalutamide in the Second Hospital of Tianjin Medical University from January 2022 to March 2023. The primary end points were the incidence of skin rash and the time to skin rash. The second end points were the grade of skin rash, the time to remission, the rate of recurrence of skin rash, clinical risk factors and management of skin rash. RESULTS: One hundred patients were analyzed. Patients were a median of 73 years old (IQR 68-77.75). Thirty-two patients (32%) developed apalutamide­associated skin rash. The median time to incidence and remission of skin rash were 57.5 and 11.5 days, respectively. Of 32 skin rash, 27 patients had apalutamide therapy maintained after rash remission. There were seven patients having recurrence of skin rash. By multivariable logistic regression analysis, we revealed that hypertension history (OR 3.22, 95% CI 1.09-9.53, p = 0.035), bad life-styles (OR 3.29, 95% CI 1.11-9.8, p = 0.032), ECOG ≥ 1 (OR 3.92, 95% CI 1.33-11.55, p = 0.013), and high tumor burden (OR 3.13, 95% CI 1.07-9.14, p = 0.037) were independently associated with higher incidence of skin rash. CONCLUSION: Nearly one-third of Chinese patients experienced skin rash after taking apalutamide in our study. The poor health patients might have a higher incidence of apalutamide-associated skin rash.

UAV hyperspectral analysis of secondary salinization in arid oasis cotton fields: effects of FOD feature selection and SOA-RF.

Secondary salinization is a crucial constraint on agricultural progress in arid regions. The specific mulching irrigation technique not only exacerbates secondary salinization but also complicates field-scale soil salinity monitoring. UAV hyperspectral remote sensing offers a monitoring method that is high-precision, high-efficiency, and short-cycle. In this study, UAV hyperspectral images were used to derive one-dimensional, textural, and three-dimensional feature variables using Competitive adaptive reweighted sampling (CARS), Gray-Level Co-occurrence Matrix (GLCM), Boruta Feature Selection (Boruta), and Brightness-Color-Index (BCI) with Fractional-order differentiation (FOD) processing. Additionally, three modeling strategies were developed (Strategy 1 involves constructing the model solely with the 20 single-band variable inputs screened by the CARS algorithm. In Strategy 2, 25 texture features augment Strategy 1, resulting in 45 feature variables for model construction. Strategy 3, building upon Strategy 2, incorporates six triple-band indices, totaling 51 variables used in the model's construction) and integrated with the Seagull Optimization Algorithm for Random Forest (SOA-RF) models to predict soil electrical conductivity (EC) and delineate spatial distribution. The results demonstrated that fractional order differentiation highlights spectral features in noisy spectra, and different orders of differentiation reveal different hidden information. The correlation between soil EC and spectra varies with the order. 1.9th order differentiation is proved to be the best order for constructing one-dimensional indices; although the addition of texture features slightly improves the accuracy of the model, the integration of the three-waveband indices significantly improves the accuracy of the estimation, with an R2 of 0.9476. In contrast to the conventional RF model, the SOA-RF algorithm optimizes its parameters thereby significantly improving the accuracy and model stability. The optimal soil salinity prediction model proposed in this study can accurately, non-invasively and rapidly identify excessive salt accumulation in drip irrigation under membrane. It is of great significance to improve the growing conditions of cotton, increase the cotton yield, and promote the sustainable development of Xinjiang's agricultural economy, and also provides a reference for the prevention and control of regional soil salinization.

Based on PI-RADS v2.1 combining PHI and ADC values to guide prostate biopsy in patients with PSA 4-20 ng/mL.

PURPOSE: The study aimed to investigate the diagnostic accuracy of prostate health index (PHI) and apparent diffusion coefficient (ADC) values in predicting prostate cancer (PCa) and construct a nomogram for the prediction of PCa and clinically significant PCa (CSPCa) in Prostate Imaging-Reporting and Data System (PI-RADS) three lesions cohort. METHODS: This study prospectively enrolled 301 patients who underwent multiparametric magnetic resonance (mpMRI) and were scheduled for prostate biopsy. The receiver operating characteristic curve (ROC) was performed to estimate the diagnostic accuracy of each predictor. Univariable and multivariable logistic regression analysis was conducted to ascertain hidden risk factors and constructed nomograms in PI-RADS three lesions cohort. RESULTS: In the whole cohort, the area under the ROC curve (AUC) of PHI is relatively high, which is 0.779. As radiographic parameters, the AUC of PI-RADS and ADC values was 0.702 and 0.756, respectively. The utilization of PHI and ADC values either individually or in combination significantly improved the diagnostic accuracy of the basic model. In PI-RADS three lesions cohort, the AUC for PCa was 0.817 in the training cohort and 0.904 in the validation cohort. The AUC for CSPCa was 0.856 in the training cohort and 0.871 in the validation cohort. When applying the nomogram for predicting PCa, 50.0% of biopsies could be saved, supplemented by 6.9% of CSPCa being missed. CONCLUSION: PHI and ADC values can be used as predictors of CSPCa. The nomogram included PHI, ADC values and other clinical predictors demonstrated an enhanced capability in detecting PCa and CSPCa within PI-RADS three lesions cohort.

Collagen Anchoring Protein-Nucleic Acid Chimeric Probe for In Situ In Vivo Mapping of a Tumor-Specific Protease.

In situ analysis of biomarkers in the tumor microenvironment (TME) is important to reveal their potential roles in tumor progression and early diagnosis of tumors but remains a challenge. In this work, a bottom-up modular assembly strategy was proposed for a multifunctional protein-nucleic chimeric probe (PNCP) for in situ mapping of cancer-specific proteases. PNCP, containing a collagen anchoring module and a target proteolysis-responsive isothermal amplification sensor module, can be anchored in the collagen-rich TME and respond to the target protease in situ and generate amplified signals through rolling cycle amplification of tandem fluorescent RNAs. Taking matrix metalloproteinase 2 (MMP-2), a tumor-associated protease, as the model, the feasibility of PNCP was demonstrated for the in situ detection of MMP-2 activity in 3D tumor spheroids. Moreover, in situ in vivo mapping of MMP-2 activity was also achieved in a metastatic solid tumor model with high sensitivity, providing a useful tool for evaluating tumor metastasis and distinguishing highly aggressive forms of tumors.

A collagen-immobilized nanodevice for in situ ratiometric imaging of cancer biomarkers in the tumor microenvironment.

Monitoring the spatiotemporal dynamics of cancer biomarkers within the tumor microenvironment (TME) is critical to understanding their roles in tumorigenesis. Here, we reported a multifunctional fusion protein (collagen-binding domain and duck circovirus tag fused to mCherry, CBD-mCherry-DCV) capable of binding collagen with high affinity and covalently binding specific nucleic acids with exceptional efficiency. We then constructed a chimeric protein-nucleic acid nanodevice (CPNN) using CBD-mCherry-DCV and an aptamer-based sensing module to enable spatially controlled ratiometric imaging of cancer biomarkers in the TME. The collagen-anchoring module CBD-mCherry-DCV allowed specific immobilization of CPNN on 3D multicellular tumor spheroids, enabling the sensing module to achieve "off-on" fluorescence imaging of cancer biomarkers upon specific target recognition by an aptamer. Taking advantage of the constant fluorescence signal of mCherry and the activatable fluorescence response of Cy5 to specific cancer biomarkers, the detection sensitivity and reliability of CPNN were improved by self-calibrating the signal intensity. Specifically, CPNN enabled ratiometric fluorescence imaging of varying concentrations of exogenous PDGF-BB and ATP in tumor spheroids with a high signal-to-background ratio. Furthermore, it allowed the visual monitoring of endogenous PDGF-BB and ATP released from cells. Overall, this study demonstrates the potential of the nanodevice as a versatile approach for the visualization and imaging of cancer biomarkers in the TME.

Long-Term Prognosis of Different Subtypes of Left Ventricular Noncompaction Cardiomyopathy Patients: A Retrospective Study in China.

Left ventricular noncompaction (LVNC) is a heterogeneous cardiomyopathy that can be classified into different subtypes based on morphologic and functional features. However, the prognosis of the dilated and isolated subtypes of non-pediatric LVNC remains unknown. We retrospectively studied 101 patients with LVNC diagnosed at Peking Union Medical College Hospital from 2006 to 2022 using the Jenni criteria of transthoracic echocardiography. The patients were grouped into those with dilated LVNC (n = 64) or isolated LVNC (n = 37), and 88 patients (54 with dilated LVNC and 34 with isolated LVNC) were followed up successfully. The primary outcome was major adverse cardiovascular events (a composite of cardiovascular mortality, heart failure, severe ventricular arrhythmia, and systolic embolism). The median follow-up time was 5.24 years. The incidence of major adverse cardiovascular events was 43.2%; patients with dilated LVNC had a higher risk (adjusted hazard ratio, 4.43; 95% confidence interval, 1.24-15.81; p = 0.02) than those with isolated LVNC. None of the isolated LVNC patients had cardiovascular deaths or severe ventricular arrhythmias. The risk of systemic embolism was similar between patients with dilated and isolated LVNC. Our findings indicate that transthoracic echocardiography is a useful tool for classifying LVNC into subtypes with distinct clinical outcomes. Dilated LVNC is associated with a poor prognosis, while the isolated subtype is probably a physiological condition.

MYC overexpression but not MYC/BCL2 double expression predicts survival in bulky mass diffuse large B-cell lymphoma patients.

PURPOSE: The prognostic factors for diffuse large B-cell lymphoma (DLBCL) have been fully explored, but prognostic information for bulky mass DLBCL patients is limited. This study aimed to analyze the prognostic value of MYC protein expression and other biological parameters in bulky mass DLBCL patients. METHODS: We defined a bulky mass as a maximum tumor diameter ≥7.5 cm and studied 227 patients with de novo bulky mass DLBCL. RESULTS: In all patients with bulky mass DLBCL, the 1-year and 3-year OS rates were 72.7% and 57.1%, respectively, and the 1-year and 3-year PFS rates were 52.0% and 42.5%, respectively. The MYC overexpression group (n = 140) showed significantly worse overall survival (OS; p = 0.019) and progression-free survival (PFS; p = 0.001) than the non-MYC overexpression group (n = 87). Subgroup analyses demonstrated that the MYC overexpression group was associated with inferior OS and PFS in the subgroups with the International Prognostic Index score of 3-5 (OS: p = 0.011; PFS: p < 0.001), Ann Arbor stage 3-4 (OS: p = 0.014; PFS: p < 0.001) and GCB subtype (OS: p = 0.014; PFS: p = 0.010). Consolidation radiotherapy improved OS and PFS in patients with bulky mass DLBCL (OS: p = 0.008; PFS: p = 0.004) as well as in those with MYC overexpression (OS: p = 0.001; PFS: p = 0.001). The prognostic value of MYC overexpression was maintained in a multivariate model adjusted for the International Prognostic Index. CONCLUSION: MYC overexpression is a poor predictor for bulky mass DLBCL patients. Consolidation radiotherapy for residual disease after induction therapy may improve outcomes for patients with bulky mass DLBCL.

SUV39H1 is a prognosis and immune microenvironment-related biomarker in diffuse large B-cell lymphoma

Background The tumor microenvironment plays a crucial role in the oncogenesis and treatment of diffuse large B-cell lymphoma (DLBCL). The H3K9me3-specific histone methyltransferase Suppressor of variegation 3-9 homolog 1 (SUV39H1) is a significant gene that promotes the progression of various malignancies. However, the specific expression of SUV39H1 in DLBCL remains unclear. Methods By retrieving data from GEPIA, UCSC XENA and TCGA public databases, we observed the high expression of SUV39H1 in DLBCL. Combined with an immunohistochemical validation assay, we analyzed our hospital’s clinical characteristics and prognosis of 67 DLBCL patients. The results showed that high SUV39H1 expression was closely associated with age over 50 years (P = 0.014) and low albumin levels (P = 0.023) of patients. Furthermore, the experiments in vitro were deployed to evaluate the regulation of SUV39H1 on the DLBCL immune microenvironment. Results The results showed that high SUV39H1 expression was closely associated with age over 50 years (P = 0.014) and low albumin levels (P = 0.023) of patients. The prognostic analysis showed that the high SUV39H1 expression group had a lower disease-free survival (DFS) rate than the low SUV39H1 expression group (P < 0.05). We further discovered that SUV39H1 upregulated the expression of CD86+ and CD163+ tumor-associated macrophages by DLBCL patients’ tissues and cell experiments in vitro (P < 0.05). And SUV39H1-associated T lymphocyte subsets and cytokines IL-6/CCL-2 were downregulated in DLBCL (P < 0.05). Conclusions In summary, SUV39H1 might be not only a potential target for treating DLBCL but also a clinical indicator for doctors to evaluate the trend of disease development (AU)

Genome-wide association analysis and Mendelian randomization proteomics identify drug targets for heart failure.

We conduct a large-scale meta-analysis of heart failure genome-wide association studies (GWAS) consisting of over 90,000 heart failure cases and more than 1 million control individuals of European ancestry to uncover novel genetic determinants for heart failure. Using the GWAS results and blood protein quantitative loci, we perform Mendelian randomization and colocalization analyses on human proteins to provide putative causal evidence for the role of druggable proteins in the genesis of heart failure. We identify 39 genome-wide significant heart failure risk variants, of which 18 are previously unreported. Using a combination of Mendelian randomization proteomics and genetic cis-only colocalization analyses, we identify 10 additional putatively causal genes for heart failure. Findings from GWAS and Mendelian randomization-proteomics identify seven (CAMK2D, PRKD1, PRKD3, MAPK3, TNFSF12, APOC3 and NAE1) proteins as potential targets for interventions to be used in primary prevention of heart failure.

A prediction model for severe hematological toxicity of BTK inhibitors.

Bruton's tyrosine kinase inhibitor (BTKi) has revolutionized the treatment of B-cell lymphomas. However, BTKi-related hematological toxicity hinders treatment continuity and may further affect clinical efficacy. To identify risk factors and predict the likelihood of BTKi-related hematological toxicities, we constructed and validated a prediction model for severe hematological toxicity of BTKi. Approved by the hospital medical science research ethics committee (No. M2022427), we collected real-world data in patients treated with BTKi from a Lymphoma Research Center in China. The outcome of interest was severe hematological toxicity caused by BTKi. 36 candidate variables were categorized into demographics, diagnostic and treatment information, laboratory data, and medical history. The study sample was randomly divided into training (70%) and validation (30%) sets. We developed and compared the performance of various modelling methods, including decision tree (DT), random forest (RF), gradient boosting decision tree (GBDT), extreme gradient boosting (XGBoost), light gradient boosting machine (LightGBM), and logistic regression (LR). Finally, we constructed a Web-calculator of the optimal model to estimate the risk of hematological toxicity. This study was designed, conducted and reported strictly in compliance with the TRIPOD checklist. Data from a total 121 patients were included [median age, 65 years (range, 56-73 years); 74 (61.15%) men; 47 (38.84%) severe hematological toxicity]. The XGBoost model demonstrated better overall properties than other models, achieving high discrimination (AUC: 0.671; accuracy: 0.730; specificity: 0.913) and clinical benefit. The following 10 variables were used to develop the XGBoost model: white blood cell count (WBC), neutrophil count (Neut), red blood cell count (RBC), platelet count (PLT), fibrinogen (Fib), total protein (TP), aspartate aminotransferase (AST), lactate dehydrogenase (LDH), gender and type of BTKi. SHAP values demonstrated insightful associations between these variables and hematological toxicity. Finally, to facilitate clinical and research use, we also deploy the XGBoost model on a web-calculator for free access. The XGBoost model with promising accuracy was developed to predict the severe hematological toxicity of BTKi. It helps to strengthen the proactive monitoring and management of patients with hematological toxicity, and thus achieve long-term continuous BTKi treatment.