Schistosome Infection Impacts Hematopoiesis.

Helminth infections are common in animals. However, the impact of a helminth infection on the function of hematopoietic stem cells (HSCs) and other hematopoietic cells has not been comprehensively defined. In this article, we describe the hematopoietic response to infection of mice with Schistosoma mansoni, a parasitic flatworm that causes schistosomiasis. We analyzed the frequency or number of hematopoietic cell types in the bone marrow, spleen, liver, thymus, and blood and observed multiple hematopoietic changes caused by infection. Schistosome infection impaired bone marrow HSC function after serial transplantation. Functional HSCs were present in the infected liver. Infection blocked bone marrow erythropoiesis and augmented spleen erythropoiesis, observations consistent with the anemia and splenomegaly prevalent in schistosomiasis patients. This work defines the hematopoietic response to schistosomiasis, a debilitating disease afflicting more than 200 million people, and identifies impairments in HSC function and erythropoiesis.

Structural variation cooperates with permissive chromatin to control enhancer hijacking-mediated oncogenic transcription.

Structural variants (SVs) involving enhancer hijacking can rewire chromatin topologies to cause oncogene activation in human cancers, including hematologic malignancies; however, because of the lack of tools to assess their effects on gene regulation and chromatin organization, the molecular determinants for the functional output of enhancer hijacking remain poorly understood. Here, we developed a multimodal approach to integrate genome sequencing, chromosome conformation, chromatin state, and transcriptomic alteration for quantitative analysis of transcriptional effects and structural reorganization imposed by SVs in leukemic genomes. We identified known and new pathogenic SVs, including recurrent t(5;14) translocations that cause the hijacking of BCL11B enhancers for the allele-specific activation of TLX3 in a subtype of pediatric leukemia. Epigenetic perturbation of SV-hijacked BCL11B enhancers impairs TLX3 transcription, which are required for the growth of t(5;14) leukemia cells. By CRISPR engineering of patient-derived t(5;14) in isogenic leukemia cells, we uncovered a new mechanism whereby the transcriptional output of SV-induced BCL11B enhancer hijacking is dependent on the loss of DNA hypermethylation at the TLX3 promoter. Our results highlight the importance of the cooperation between genetic alteration and permissive chromatin as a critical determinant of SV-mediated oncogene activation, with implications for understanding aberrant gene transcription after epigenetic therapies in patients with leukemia. Hence, leveraging the interdependency of genetic alteration on chromatin variation may provide new opportunities to reprogram gene regulation as targeted interventions in human disease.

YAP activation in liver macrophages via depletion of MST1/MST2 enhances liver inflammation and fibrosis in MASLD.

Macrophages have been recognized as pivotal players in the progression of MASLD/MASH. However, the molecular mechanisms underlying their multifaceted functions in the disease remain to be further clarified. In the current study, we developed a new mouse model with YAP activation in macrophages to delineate the effect and mechanism of YAP signaling in the pathogenesis of MASLD/MASH. Genetically modified mice, featuring specific depletion of both Mst1 and Mst2 in macrophages/monocytes, were generated and exposed to a high-fat diet for 12 weeks to induce MASLD. Following this period, livers were collected for histopathological examination, and liver non-parenchymal cells were isolated and subjected to various analyses, including single-cell RNA-sequencing, immunofluorescence and immunoblotting and qRT-PCR to investigate the impact of YAP signaling on the progression of MASLD. Our data revealed that Mst1/2 depletion in liver macrophages enhanced liver inflammation and fibrosis in MASLD. Using single-cell RNA-sequencing, we showed that YAP activation via Mst1/2 depletion upregulated the expressions of both pro-inflammatory genes and genes associated with resolution/tissue repair. We observed that YAP activation increases Kupffer cell populations (i.e., Kupffer-2 and Kupffer-3) which are importantly implicated in the pathogenesis of MASLD/MASH. Our data indicate that YAP activation via Mst1/2 deletion enhances both the pro-inflammatory and tissue repairing functions of Kupffer-1 and -2 cells at least in part through C1q. These YAP-regulatory mechanisms control the plasticity of liver macrophages in the context of MASLD/MASH. Our findings provide important evidence supporting the critical regulatory role of YAP signaling in liver macrophage plasticity and the progression of MASLD. Therefore, targeting the Hippo-YAP pathway may present a promising therapeutic strategy for the treatment of MASH.

T/myeloid mixed phenotype acute leukaemia harbouring TLX3::BCL11B with TLX3 activation.

T/myeloid mixed phenotype acute leukaemia (MPAL) is a rare aggressive acute leukaemia with poorly understood pathogenesis. Herein, we report two cases of T/myeloid MPAL harbouring BCL11B-associated structural variants that activate TLX3 (TLX3::BCL11B-TLX3-activation) by genome sequencing and transcriptomic analyses. Both patients were young males with extramedullary involvement. Cooperative gene alterations characteristic of T/myeloid MPAL and T-lymphoblastic leukaemia (T-ALL) were detected. Both patients achieved initial remission following lineage-matched ALL-based therapy with one patient requiring a lineage-switched myeloid-based therapy. Our study is the first to demonstrate the clinicopathological and genomic features of TLX3::BCL11B-TLX3-activated T/myeloid MPAL and provide insights into leukaemogenesis.

A novel imprinted locus on bovine chromosome 18 homologous with human chromosome 16q24.1.

Genomic imprinting is an epigenetic regulation mechanism in mammals resulting in the parentally dependent monoallelic expression of genes. Imprinting disorders in humans are associated with several congenital syndromes and cancers and remain the focus of many medical studies. Cattle is a better model organism for investigating human embryo development than mice. Imprinted genes usually cluster on chromosomes and are regulated by different methylation regions (DMRs) located in imprinting control regions that control gene expression in cis. There is an imprinted locus on human chromosome 16q24.1 associated with congenital lethal developmental lung disease in newborns. However, genomic imprinting on bovine chromosome 18, which is homologous with human chromosome 16 has not been systematically studied. The aim of this study was to analyze the allelic expressions of eight genes (CDH13, ATP2C2, TLDC1, COTL1, CRISPLD2, ZDHHC7, KIAA0513, and GSE1) on bovine chromosome 18 and to search the DMRs associated gene allelic expression. Three transcript variants of the ZDHHC7 gene (X1, X2, and X5) showed maternal imprinting in bovine placentas. In addition, the monoallelic expression of X2 and X5 was tissue-specific. Five transcripts of the KIAA0513 gene showed tissue- and isoform-specific monoallelic expression. The CDH13, ATP2C2, and TLDC1 genes exhibited tissue-specific imprinting, however, COTL1, CRISLPLD2, and GSE1 escaped imprinting. Four DMRs, established after fertilization, were found in this region. Two DMRs were located between the ZDHHC7 and KIAA0513 genes, and two were in exon 1 of the CDH13 and ATP2C2 genes, respectively. The results from this study support future studies on the molecular mechanism to regulate the imprinting of candidate genes on bovine chromosome 18.

Outcomes in Kaposi's sarcoma-associated herpesvirus -associated primary effusion lymphoma and multicentric Castleman's disease in patients with human immunodeficiency virus (HIV) in a safety-net hospital system.

OBJECTIVE: To describe cases of Kaposi's sarcoma-associated herpesvirus (KSHV)-associated multicentric Castleman's disease (MCD) and primary effusion lymphoma (PEL) in patients with HIV from a large, safety-net hospital system in Dallas, Texas, USA. METHODS: We conducted a retrospective review of patients with HIV-associated PEL and/or MCD. RESULTS: Twelve patients with PEL and 10 patients with MCD were identified. All patients were male and 17 of 20 were men who have sex with men; 66.7% of PEL patients and 50% of MCD patients had concurrent KS at the time of diagnosis; 42% of patients with PEL and 20% of patients with MCD died during the follow-up period. We noted improved survival in our cohort compared to previous studies, particularly in our PEL patients with a median survival of 11.4 months compared to 3-6-month median survival historically. Median follow-up time for MCD patients was 17.5 months. This improved survival is despite suboptimal antiretroviral therapy (ART) adherence at diagnosis, with only 50% of patients on ART at the time of MCD/PEL diagnosis. CONCLUSION: These data highlight the importance of early recognition of PEL and MCD, and the larger-scale efforts needed to better understand the pathogenetic drivers of clinical outcomes in patients affected by KSHV-related diseases.

LILRB4 signalling in leukaemia cells mediates T cell suppression and tumour infiltration.

Immune checkpoint blockade therapy has been successful in treating some types of cancer but has not shown clinical benefits for treating leukaemia1. This result suggests that leukaemia uses unique mechanisms to evade this therapy. Certain immune inhibitory receptors that are expressed by normal immune cells are also present on leukaemia cells. Whether these receptors can initiate immune-related primary signalling in tumour cells remains unknown. Here we use mouse models and human cells to show that LILRB4, an immunoreceptor tyrosine-based inhibition motif-containing receptor and a marker of monocytic leukaemia, supports tumour cell infiltration into tissues and suppresses T cell activity via a signalling pathway that involves APOE, LILRB4, SHP-2, uPAR and ARG1 in acute myeloid leukaemia (AML) cells. Deletion of LILRB4 or the use of antibodies to block LILRB4 signalling impeded AML development. Thus, LILRB4 orchestrates tumour invasion pathways in monocytic leukaemia cells by creating an immunosuppressive microenvironment. LILRB4 represents a compelling target for the treatment of monocytic AML.

Exposure to high dose of polystyrene nanoplastics causes trophoblast cell apoptosis and induces miscarriage.

BACKGROUND: With rapid increase in the global use of various plastics, microplastics (MPs) and nanoplastics (NPs) pollution and their adverse health effects have attracted global attention. MPs have been detected out in human body and both MPs and NPs showed female reproductive toxicological effects in animal models. Miscarriage (abnormal early embryo loss), accounting for 15-25% pregnant women worldwide, greatly harms human reproduction. However, the adverse effects of NPs on miscarriage have never been explored. RESULTS: In this study, we identified that polystyrene (PS) plastics particles were present in women villous tissues. Their levels were higher in villous tissues of unexplained recurrent miscarriage (RM) patients vs. healthy control (HC) group. Furthermore, mouse assays further confirmed that exposure to polystyrene nanoplastics (PS-NPs, 50 nm in diameter, 50 or 100 mg/kg) indeed induced miscarriage. In mechanism, PS-NPs exposure (50, 100, 150, or 200 µg/mL) increased oxidative stress, decreased mitochondrial membrane potential, and increased apoptosis in human trophoblast cells by activating Bcl-2/Cleaved-caspase-2/Cleaved-caspase-3 signaling through mitochondrial pathway. The alteration in this signaling was consistent in placental tissues of PS-NPs-exposed mouse model and in villous tissues of unexplained RM patients. Supplement with Bcl-2 could efficiently suppress apoptosis in PS-NPs-exposed trophoblast cells and reduce apoptosis and alleviate miscarriage in PS-NPs-exposed pregnant mouse model. CONCLUSIONS: Exposure to PS-NPs activated Bcl-2/Cleaved-caspase-2/Cleaved-caspase-3, leading to excessive apoptosis in human trophoblast cells and in mice placental tissues, further inducing miscarriage.

The landscape of allelic expression and DNA methylation at the bovine SGCE/PEG10 locus.

Genomic imprinting is an epigenetic regulation in mammals in which a small subset of genes is monoallelically expressed dependent on their parental origin. A large imprinted domain, SGCE/PEG10 locus, is located on human chromosome 7q21s and mouse proximal chromosome 6. However, genomic imprinting of bovine SGCE/PEG10 cluster has not been systematically studied. In this study, we investigated allele expression of 14 genes of the SGCE/PEG10 locus in bovine somatic tissues and term placenta using a single nucleotide polymorphism (SNP)-based sequencing method. In addition to SGCE and PEG10, two conserved paternally expressed genes in human and mice, five other genes (TFPI2, GNG11, ASB4, PON1, and PON3) were paternally expressed. Three genes, BET1, COL1A2, and CASD1, exhibited tissue-specific monoallelic expression. CALCR showed monoallelic expression in tissues but biallelic expression in the placenta. Three genes, GNGT1, PPP1R9A, and PON2, showed biallelic expression in cattle. Five differentially methylated regions (DMRs) were found to be associated with the allelic expression of TFPI2, COL1A2, SGCE/PEG10, PON3, and ASB4 genes, respectively. The SGCE/PEG10 DMR is a maternally hypermethylated germline DMR, but TFPI2, COL1A2, PON3, and ASB4 DMRs are secondary DMRs. In summary, we identified five novel bovine imprinted genes (GNG11, BET1, COL1A2, CASD1, and PON1) and four secondary DMRs at the SGCE/PEG10 locus.

Environmental BaP/BPDE suppressed trophoblast cell invasion/migration and induced miscarriage by down-regulating lnc-HZ01/MEST/VIM axis.

Environmental benzo(a)pyrene (BaP) and itsmetabolite benzo(a)pyrene-7, 8-dihydrodiol-9, 10-epoxide (BPDE), classic endocrine disrupting chemical and persistent organic pollutant, could cause miscarriage. However, the detailed mechanisms are still largely unclear and should be further explored. In this study, we discovered that exposure of trophoblast cells with BPDE could suppressed cell invasion/migration by inhibiting MEST/VIM (Vimentin) pathway. Moreover, BPDE exposure also increased lnc-HZ01 expression level, which further inhibited MEST/VIM pathway and then suppressed invasion/migration. Knockdown of lnc-HZ01 or overexpression of MEST could efficiently rescue invasion/migration of BPDE-exposed Swan 71 cells. Furthermore, lnc-HZ01 was highly expressed and MEST/VIM were lowly expressed in recurrent miscarriage (RM) villous tissues compared with healthy control (HC) group. Finally, we also found that BaP exposure inhibited murine Mest/Vim pathway in placental tissues and induced miscarriage in BaP-exposed mice. Therefore, the regulatory mechanisms were similar in BPDE-exposed human trophoblast cells, RM villous tissues, and placental tissues of BaP-exposed mice with miscarriage, building a bridge to connect BaP/BPDE exposure, invasion/migration, and miscarriage. This study provided novel insights in the toxicological effects and molecular mechanisms of BaP/BPDE-induced miscarriage, which is helpful for better elucidating the toxicological risks of BaP/BPDE on female reproduction.

Copper exposure induces trophoblast cell cuproptosis by up-regulating lnc-HZ11.

Copper pollution has attracted global environmental concern. Widespread Cu pollution results in excessive Cu accumulation in human. Epidemiological studies and animal experiments revealed that Cu exposure might have reproductive toxicity. Cuproptosis is a recently reported Cu-dependent and programmed cell death pattern. However, the mechanism by which copper exposure might cause cell cuproptosis is largely unknown. We chose trophoblast cells as cell model and found that copper exposure causes trophoblast cell cuproptosis. In mechanism, copper exposure up-regulates lnc-HZ11 expression levels, which increases intracellular Cu2+ levels and causes trophoblast cell cuproptosis. Knockdown of lnc-HZ11 efficiently reduces intracellular Cu2+ levels and alleviate trophoblast cell cuproptosis, which could be further alleviated by co-treatment with DC or TEPA. These results discover novel toxicological effects of copper exposure and also provide potential target for protection trophoblast cells from cuproptosis in the presence of excessive copper exposure.

Automated classification of cytogenetic abnormalities in hematolymphoid neoplasms.

MOTIVATION: Algorithms for classifying chromosomes, like convolutional deep neural networks (CNNs), show promise to augment cytogeneticists' workflows; however, a critical limitation is their inability to accurately classify various structural chromosomal abnormalities. In hematopathology, recurrent structural cytogenetic abnormalities herald diagnostic, prognostic and therapeutic implications, but are laborious for expert cytogeneticists to identify. Non-recurrent cytogenetic abnormalities also occur frequently cancerous cells. Here, we demonstrate the feasibility of using CNNs to accurately classify many recurrent cytogenetic abnormalities while being able to reliably detect non-recurrent, spurious abnormal chromosomes, as well as provide insights into dataset assembly, model selection and training methodology that improve overall generalizability and performance for chromosome classification. RESULTS: Our top-performing model achieved a mean weighted F1 score of 96.86% on the validation set and 94.03% on the test set. Gradient class activation maps indicated that our model learned biologically meaningful feature maps, reinforcing the clinical utility of our proposed approach. Altogether, this work: proposes a new dataset framework for training chromosome classifiers for use in a clinical environment, reveals that residual CNNs and cyclical learning rates confer superior performance, and demonstrates the feasibility of using this approach to automatically screen for many recurrent cytogenetic abnormalities while adeptly classifying non-recurrent abnormal chromosomes. AVAILABILITY AND IMPLEMENTATION: Software is freely available at https://github.com/DaehwanKimLab/Chromosome-ReAd. The data underlying this article cannot be shared publicly due to it being protected patient information. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

EZH2 Promotes Cholangiocarcinoma Development and Progression through Histone Methylation and microRNA-Mediated Down-Regulation of Tumor Suppressor Genes.

Cholangiocarcinoma (CCA) is a highly malignant cancer of the biliary tree. Although studies have implicated enhancer of Zeste homolog 2 (EZH2) in CCA growth, the role of EZH2 in CCA development has not been investigated, and the mechanism for EZH2-regulated gene expression in CCA remains to be further defined. The current study used a mouse model of CCA induced by hydrodynamic tail vein injection of Notch1 intracellular domain and myristoylated-AKT plasmids. Mice with liver-specific EZH2 knockout displayed reduced CCA development. In a xenograft model, EZH2 knockdown significantly decreased CCA progression. Administration of the EZH2 inhibitor GSK126 decreased CCA tumor burden in mice. Accordingly, EZH2 depletion or inhibition reduced the growth and colony formation capability of CCA cells. Analysis of high-throughput data identified a set of 12 tumor-inhibiting genes as targets of EZH2 in CCA. The experimental results suggest that EZH2 may down-regulate these tumor-inhibiting genes through methylation of lysine 27 on histone H3 (H3K27) in the gene louses and through regulation of specific miRNAs. High mobility group box 1 was shown to facilitate the methyltransferase activity of EZH2, which is implicated in the regulation of CCA cell growth. The study shows that EZH2 promotes CCA development and progression through a complicated regulatory network involving tumor-inhibiting genes, miRNAs, and high mobility group box 1, which support targeting EZH2 as a potentially effective strategy for CCA treatment.

Role of tRNA-derived small RNAs(tsRNAs) in the diagnosis and treatment of malignant tumours.

Malignant tumours area leading cause of death globally, accounting for approximately 13% of all deaths. A detailed understanding of the mechanism(s) of the occurrence and development of malignant tumours and identification of relevant therapeutic targets are therefore key to tumour treatment. tsRNAs(tRNA-derived small RNAs)-also known as TRFs (tRNA-derived fragments), tiRNAs (tRNA-derived stress-induced RNAs), tRNA halves, etc.-are a recently identified class of small noncoding RNAs that are generated from mature tRNA or tRNA precursors through cleavage by enzymes such as angiogenin, Dicer, RNase Z, and RNase P. Several studies have confirmed that dysregulation of tsRNAs is closely related to the tumorigenesis of breast cancer, nasopharyngeal cancer, lung cancer, and so on. Furthermore, research indicates that tsRNAs can be used as clinical diagnostic markers and therapeutic targets for cancer. In our review, we summarized the recent research progress on the role and clinical application of tsRNAs in tumorigenesis and progression. Video Abstract.

LKB1 signaling and patient survival outcomes in hepatocellular carcinoma.

The liver is a major organ that is involved in essential biological functions such as digestion, nutrient storage, and detoxification. Furthermore, it is one of the most metabolically active organs with active roles in regulating carbohydrate, protein, and lipid metabolism. Hepatocellular carcinoma is a cancer of the liver that is associated in settings of chronic inflammation such as viral hepatitis, repeated toxin exposure, and fatty liver disease. Furthermore, liver cancer is the most common cause of death associated with cirrhosis and is the 3rd leading cause of global cancer deaths. LKB1 signaling has been demonstrated to play a role in regulating cellular metabolism under normal and nutrient deficient conditions. Furthermore, LKB1 signaling has been found to be involved in many cancers with most reports identifying LKB1 to have a tumor suppressive role. In this review, we use the KMPlotter database to correlate RNA levels of LKB1 signaling genes and hepatocellular carcinoma patient survival outcomes with the hopes of identifying potential biomarkers clinical usage. Based on our results STRADß, CAB39L, AMPKα, MARK2, SIK1, SIK2, BRSK1, BRSK2, and SNRK expression has a statistically significant impact on patient survival.

Identifying patients at risk for hereditary myeloid malignancy syndromes incorporating a novel, self-administered questionnaire to an initial screening platform.

INTRODUCTION: Four to 10% of cases of myeloid malignancies are inherited. We report our experience on hereditary myeloid malignancy syndromes (HMMS) incorporating a novel questionnaire in the screening platform for patients with myeloid malignancies and aplastic anemia. METHODS: The questionnaire was sent via electronic patient portal prior to clinic visits. Patients screened positive based on responses to questionnaire items, presence of suspicion disease characteristics (young age, family history, monosomy 7 etc.) and/or presence of signs of HMMS. Those deemed at-risk based on questionnaire responses, clinical features and/or somatic mutation profile were offered germline testing. RESULTS: A total of 408 patients were screened, 141 (35%) were deemed at-risk. Fifty-four (38%) of at-risk patients were seen in the genetics clinic. Forty-one (76%) of the patients seen agreed to germline testing and 13 declined due to cost or personal decision. Twenty pathogenic (P)/likely-pathogenic (LP) germline mutations were identified in 16 (39%) of the tested patients. Five patients also had a variant of uncertain significance (VUS) and an additional 13 had at least 1 VUS without P/LP mutations (total 29 VUS's were found in 18 (44%) of tested patients). The median age of diagnosis for patients with P/LP mutations was 56 years versus 66 years in the entire cohort. CONCLUSION: Incorporating an electronic questionnaire is an effective screening method for HMMS. Many patients declined testing due to cost. These results highlight the importance of germline testing in patients with myeloid malignancies, further research in HMMS, and coverage by healthcare plans.

Ascorbate regulates haematopoietic stem cell function and leukaemogenesis.

Stem-cell fate can be influenced by metabolite levels in culture, but it is not known whether physiological variations in metabolite levels in normal tissues regulate stem-cell function in vivo. Here we describe a metabolomics method for the analysis of rare cell populations isolated directly from tissues and use it to compare mouse haematopoietic stem cells (HSCs) to restricted haematopoietic progenitors. Each haematopoietic cell type had a distinct metabolic signature. Human and mouse HSCs had unusually high levels of ascorbate, which decreased with differentiation. Systemic ascorbate depletion in mice increased HSC frequency and function, in part by reducing the function of Tet2, a dioxygenase tumour suppressor. Ascorbate depletion cooperated with Flt3 internal tandem duplication (Flt3ITD) leukaemic mutations to accelerate leukaemogenesis, through cell-autonomous and possibly non-cell-autonomous mechanisms, in a manner that was reversed by dietary ascorbate. Ascorbate acted cell-autonomously to negatively regulate HSC function and myelopoiesis through Tet2-dependent and Tet2-independent mechanisms. Ascorbate therefore accumulates within HSCs to promote Tet activity in vivo, limiting HSC frequency and suppressing leukaemogenesis.

Myeloid sarcoma: An overview.

Myeloid Sarcoma (MS) is a high grade, hematological malignancy defined as an extramedullary tumor mass of myeloid blasts with or without maturation that effaces tissue architecture. It is a highly heterogenous condition that represents a variety of myeloid neoplasms. This heterogeneity of MS, together with its rarity, have greatly hampered our understanding of the condition. Diagnosis requires tumor biopsy, which should be accompanied by bone marrow evaluation for medullary disease. It is presently recommended that MS be treated similar to AML. Additionally, ablative radiotherapy and novel targeted therapies may also be beneficial. Genetic profiling has identified recurrent genetic abnormalities including gene mutations associated with MS, supporting its etiology similar to AML. However, the mechanisms by which MS homes to specific organs is unclear. This review provides an overview of pathogenesis, pathological and genetic findings, treatment, and prognosis. Improving the management and outcomes of MS patients requires a better understanding of its pathogenesis and its response to various therapeutic approaches.

Artificial intelligence in clinical multiparameter flow cytometry and mass cytometry-key tools and progress.

There are many research studies and emerging tools using artificial intelligence (AI) and machine learning to augment flow and mass cytometry workflows. Emerging AI tools can quickly identify common cell populations with continuous improvement of accuracy, uncover patterns in high-dimensional cytometric data that are undetectable by human analysis, facilitate the discovery of cell subpopulations, perform semi-automated immune cell profiling, and demonstrate potential to automate aspects of clinical multiparameter flow cytometric (MFC) diagnostic workflow. Utilizing AI in the analysis of cytometry samples can reduce subjective variability and assist in breakthroughs in understanding diseases. Here we review the diverse types of AI that are being applied to clinical cytometry data and how AI is driving advances in data analysis to improve diagnostic sensitivity and accuracy. We review supervised and unsupervised clustering algorithms for cell population identification, various dimensionality reduction techniques, and their utilities in visualization and machine learning pipelines, and supervised learning approaches for classifying entire cytometry samples.Understanding the AI landscape will enable pathologists to better utilize open source and commercially available tools, plan exploratory research projects to characterize diseases, and work with machine learning and data scientists to implement clinical data analysis pipelines.

Machine learning prediction for COVID-19 disease severity at hospital admission.

IMPORTANCE: Early prognostication of patients hospitalized with COVID-19 who may require mechanical ventilation and have worse outcomes within 30 days of admission is useful for delivering appropriate clinical care and optimizing resource allocation. OBJECTIVE: To develop machine learning models to predict COVID-19 severity at the time of the hospital admission based on a single institution data. DESIGN, SETTING, AND PARTICIPANTS: We established a retrospective cohort of patients with COVID-19 from University of Texas Southwestern Medical Center from May 2020 to March 2022. Easily accessible objective markers including basic laboratory variables and initial respiratory status were assessed using Random Forest's feature importance score to create a predictive risk score. Twenty-five significant variables were identified to be used in classification models. The best predictive models were selected with repeated tenfold cross-validation methods. MAIN OUTCOMES AND MEASURES: Among patients with COVID-19 admitted to the hospital, severity was defined by 30-day mortality (30DM) rates and need for mechanical ventilation. RESULTS: This was a large, single institution COVID-19 cohort including total of 1795 patients. The average age was 59.7 years old with diverse heterogeneity. 236 (13%) required mechanical ventilation and 156 patients (8.6%) died within 30 days of hospitalization. Predictive accuracy of each predictive model was validated with the 10-CV method. Random Forest classifier for 30DM model had 192 sub-trees, and obtained 0.72 sensitivity and 0.78 specificity, and 0.82 AUC. The model used to predict MV has 64 sub-trees and returned obtained 0.75 sensitivity and 0.75 specificity, and 0.81 AUC. Our scoring tool can be accessed at https://faculty.tamuc.edu/mmete/covid-risk.html . CONCLUSIONS AND RELEVANCE: In this study, we developed a risk score based on objective variables of COVID-19 patients within six hours of admission to the hospital, therefore helping predict a patient's risk of developing critical illness secondary to COVID-19.