ABSTRACT
BACKGROUND AND OBJECTIVES: Gut dysbiosis that resulted from the alteration between host-microbe interaction might worsen obesity-induced systemic inflammation. Gut microbiota manipulation by supplementation of prebiotic inulin may reverse metabolic abnormalities and improve obesity. This study aimed to determine whether inulin supplementation improved intestinal microbiota and microbial functional pathways in children with obesity. METHODS: Children with obesity whose BMI above median + 2SDs were recruited to a randomized, double-blinded placebo-controlled study. The participants aged 7-15 years were assigned to inulin supplement extracted from Thai Jerusalem artichoke (intervention), maltodextrin (placebo), and dietary fiber advice groups. All participants received similar monthly conventional advice and follow-up for 6 months. Fecal samples were collected for gut microbiome analysis using 16S rRNA sequencing. Phylogenetic Investigation of Communities by Reconstruction of Unobserved States was performed to infer microbial functional pathways. RESULTS: One hundred and forty-three children with available taxonomic and functional pathway abundance profiles were evaluated. A significant increase in alpha-diversity was observed in the inulin group. Inulin supplementation substantially enhanced Bifidobacterium, Blautia, Megasphaera, and several butyrate-producing bacteria, including Agathobacter, Eubacterium coprostanoligenes, and Subdoligranulum, compared to the other groups. The inulin group showed a significant difference in functional pathways of proteasome and riboflavin metabolism. These changes correlated with clinical and metabolic outcomes exclusively in the inulin group. CONCLUSIONS: Inulin supplementation significantly promoted gut bacterial diversity and improved gut microbiota dysbiosis in children with obesity. The modulation of functional pathways by inulin suggests its potential to establish beneficial interactions between the gut microbiota and host physiology. Inulin supplementation could be a strategic treatment to restore the balance of intestinal microbiota and regulate their functions in childhood obesity.
ABSTRACT
Accurate inference of orthologous genes constitutes a prerequisite for comparative and evolutionary genomics. SonicParanoid is one of the fastest tools for orthology inference; however, its scalability and accuracy have been hampered by time-consuming all-versus-all alignments and the existence of proteins with complex domain architectures. Here, we present a substantial update of SonicParanoid, where a gradient boosting predictor halves the execution time and a language model doubles the recall. Application to empirical large-scale and standardized benchmark datasets shows that SonicParanoid2 is much faster than comparable methods and also the most accurate. SonicParanoid2 is available at https://gitlab.com/salvo981/sonicparanoid2 and https://zenodo.org/doi/10.5281/zenodo.11371108 .
Subject(s)
Machine Learning , Software , Genomics/methods , Sequence Alignment , HumansABSTRACT
The objective is to evaluate the performance of blood test results, radiomics, and a combination of the two data types on the prediction of the 24-h oxygenation support need for the Coronavirus disease 2019 (COVID-19) patients. In this retrospective cohort study, COVID-19 patients with confirmed real-time reverse transcription-polymerase chain reaction assay (RT-PCR) test results between February 2020 and August 2021 were investigated. Initial blood cell counts, chest radiograph, and the status of oxygenation support used within 24 h were collected (n = 290; mean age, 45 ± 19 years; 125 men). Radiomics features from six lung zones were extracted. Logistic regression and random forest models were developed using the clinical-only, radiomics-only, and combined data. Ten repeats of fivefold cross-validation with bootstrapping were used to identify the input features and models with the highest area under the receiver operating characteristic curve (AUC). Higher AUCs were achieved when using only radiomics features compared to using only clinical features (0.94 ± 0.03 vs. 0.88 ± 0.04). The best combined model using both radiomics and clinical features achieved highest in the cross-validation (0.95 ± 0.02) and test sets (0.96 ± 0.02). In comparison, the best clinical-only model yielded AUCs of 0.88 ± 0.04 in cross-validation and 0.89 ± 0.03 in test set. Both radiomics and clinical data can be used to predict 24-h oxygenation support need for COVID-19 patients with AUC > 0.88. Moreover, the combination of both data types further improved the performance.
Subject(s)
COVID-19 , Oxygen , Radiography, Thoracic , Humans , COVID-19/diagnostic imaging , Middle Aged , Male , Female , Retrospective Studies , Adult , Oxygen/metabolism , Radiography, Thoracic/methods , Aged , Lung/diagnostic imaging , RadiomicsABSTRACT
INTRODUCTION: Methylmalonic acidemia (MMA) is a disorder of autosomal recessive inheritance, with an estimated prevalence of 1:50,000. First-tier clinical diagnostic tests often return many false positives [five false positive (FP): one true positive (TP)]. In this work, our goal was to refine a classification model that can minimize the number of false positives, currently an unmet need in the upstream diagnostics of MMA. METHODS: We developed machine learning multivariable screening models for MMA with utility as a secondary-tier tool for false positives reduction. We utilized mass spectrometry-based features consisting of 11 amino acids and 31 carnitines derived from dried blood samples of neonatal patients, followed by additional ratio feature construction. Feature selection strategies (selection by filter, recursive feature elimination, and learned vector quantization) were used to determine the input set for evaluating the performance of 14 classification models to identify a candidate model set for an ensemble model development. RESULTS: Our work identified computational models that explore metabolic analytes to reduce the number of false positives without compromising sensitivity. The best results [area under the receiver operating characteristic curve (AUROC) of 97%, sensitivity of 92%, and specificity of 95%] were obtained utilizing an ensemble of the algorithms random forest, C5.0, sparse linear discriminant analysis, and autoencoder deep neural network stacked with the algorithm stochastic gradient boosting as the supervisor. The model achieved a good performance trade-off for a screening application with 6% false-positive rate (FPR) at 95% sensitivity, 35% FPR at 99% sensitivity, and 39% FPR at 100% sensitivity. CONCLUSIONS: The classification results and approach of this research can be utilized by clinicians globally, to improve the overall discovery of MMA in pediatric patients. The improved method, when adjusted to 100% precision, can be used to further inform the diagnostic process journey of MMA and help reduce the burden for patients and their families.
ABSTRACT
BACKGROUND: The prognosis of nasopharyngeal carcinoma (NPC) is challenging due to late-stage identification and frequently undetectable Epstein-Barr virus (EBV) DNA. Incorporating radiomic features, which quantify tumor characteristics from imaging, may enhance prognosis assessment. PURPOSE: To investigate the predictive power of radiomic features on overall survival (OS), progression-free survival (PFS), and distant metastasis-free survival (DMFS) in NPC. MATERIALS AND METHODS: A retrospective analysis of 183 NPC patients treated with chemoradiotherapy from 2010 to 2019 was conducted. All patients were followed for at least three years. The pretreatment CT images with contrast medium, MR images (T1W and T2W), as well as gross tumor volume (GTV) contours, were used to extract radiomic features using PyRadiomics v.2.0. Robust and efficient radiomic features were chosen using the intraclass correlation test and univariate Cox proportional hazard regression analysis. They were then combined with clinical data including age, gender, tumor stage, and EBV DNA level for prognostic evaluation using Cox proportional hazard regression models with recursive feature elimination (RFE) and were optimized using 20 repetitions of a five-fold cross-validation scheme. RESULTS: Integrating radiomics with clinical data significantly enhanced the predictive power, yielding a C-index of 0.788 ± 0.066 to 0.848 ± 0.079 for the combined model versus 0.745 ± 0.082 to 0.766 ± 0.083 for clinical data alone (p<0.05). Multimodality radiomics combined with clinical data offered the highest performance. Despite the absence of EBV DNA, radiomics integration significantly improved survival predictions (C-index ranging from 0.770 ± 0.070 to 0.831 ± 0.083 in combined model versus 0.727 ± 0.084 to 0.734 ± 0.088 in clinical model, p<0.05). CONCLUSIONS: The combination of multimodality radiomic features from CT and MR images could offer superior predictive performance for OS, PFS, and DMFS compared to relying on conventional clinical data alone.
Subject(s)
Epstein-Barr Virus Infections , Nasopharyngeal Neoplasms , Humans , Nasopharyngeal Carcinoma/pathology , Epstein-Barr Virus Infections/pathology , Retrospective Studies , Nasopharyngeal Neoplasms/diagnostic imaging , Nasopharyngeal Neoplasms/therapy , Nasopharyngeal Neoplasms/pathology , Radiomics , Herpesvirus 4, Human/genetics , Prognosis , DNA , DNA, ViralABSTRACT
BACKGROUND: Diagnosing drug-induced allergy, especially nonimmediate phenotypes, is challenging. Incorrect classifications have unwanted consequences. OBJECTIVE: We sought to evaluate the diagnostic utility of IFN-γ ELISpot and clinical parameters in predicting drug-induced nonimmediate hypersensitivity using machine learning. METHODS: The study recruited 393 patients. A positive patch test or drug provocation test (DPT) was used to define positive drug hypersensitivity. Various clinical factors were considered in developing random forest (RF) and logistic regression (LR) models. Performances were compared against the IFN-γ ELISpot-only model. RESULTS: Among the 102 patients who had 164 DPTs, most patients had severe cutaneous adverse reactions (35/102, 34.3%) and maculopapular exanthems (33/102, 32.4%). Common suspected drugs were antituberculosis drugs (46/164, 28.1%) and ß-lactams (42/164, 25.6%). Mean (SD) age of patients with DPT was 52.7 (20.8) years. IFN-γ ELISpot, fixed drug eruption, Naranjo categories, and nonsteroidal anti-inflammatory drugs were the most important features in all developed models. The RF and LR models had higher discriminating abilities. An IFN-γ ELISpot cutoff value of 16.0 spot-forming cells/106 PBMCs achieved 94.8% specificity and 57.1% sensitivity. Depending on clinical needs, optimal cutoff values for RF and LR models can be chosen to achieve either high specificity (0.41 for 96.1% specificity and 0.52 for 97.4% specificity, respectively) or high sensitivity (0.26 for 78.6% sensitivity and 0.37 for 71.4% sensitivity, respectively). CONCLUSIONS: IFN-γ ELISpot assay was valuable in identifying culprit drugs, whether used individually or incorporated in a prediction model. Performances of RF and LR models were comparable. Additional test datasets with DPT would be helpful to validate the model further.
Subject(s)
Drug Hypersensitivity , Humans , Middle Aged , Drug Hypersensitivity/diagnosis , beta-Lactams/adverse effects , Immunologic Tests , Enzyme-Linked Immunospot Assay , Patch TestsABSTRACT
BACKGROUND: The published reference ranges for Doppler parameters of the fetal pulmonary artery (PA) are usually derived from small sample sizes with no practical standard score or percentile ranking, which hinders systematic comparisons of Doppler figures across different gestational ages (GAs). This study aimed to establish comprehensive reference ranges and provide a percentile ranking solution for key spectral Doppler parameters. METHODS: This is a cross-sectional study of 465 uncomplicated singleton pregnancies during 20 to 40 weeks of gestation. Spectral waveforms of the fetal main branch PA were obtained with a pulsed-wave Doppler interrogation site within 5 mm from the vascular origin. Fifteen spectral Doppler parameters were identified. Associations between these parameters with GA and fetal heart rate were assessed and used to develop percentile calculators via different statistical models. The root mean squared error of each model was calculated to determine the best performance solution. RESULTS: Acceptable spectral waveforms were obtained for 94.1% (438/465) of the fetuses. All Doppler parameters except pulsatility index, manually traced pulsatility index, peak systolic velocity, and time to systolic notch/acceleration time ratio were significantly correlated with GA, while acceleration time, ejection time, time to systolic notch, peak early-diastolic reversal flow, and peak early-diastolic reversal flow/peak systolic velocity ratio were additionally significantly correlated with fetal heart rate. Support vector machine models with radial basis kernel yield the best percentile estimation (root mean squared error of 2.17-4.08 and R2 of >0.98). Furthermore, the top 5% and bottom 5% outliers could be identified with positive predictive values of 0.71 to 0.97. An online user interface of percentile calculators is available at https://github.com/cmb-chula/fetoPAD. CONCLUSIONS: This study presents normal reference ranges and percentile calculators for 15 spectral Doppler parameters of the fetal main branch PA, some of which have not been published. The estimated percentiles enhance comparison and outlier detection of the spectral Doppler figures among fetuses at different GAs.
Subject(s)
Pulmonary Artery , Ultrasonography, Prenatal , Pregnancy , Female , Humans , Pulmonary Artery/diagnostic imaging , Reference Values , Cross-Sectional Studies , Blood Flow Velocity , Fetus/diagnostic imaging , Gestational AgeABSTRACT
MOTIVATION: The binding of a peptide antigen to a Class I major histocompatibility complex (MHC) protein is part of a key process that lets the immune system recognize an infected cell or a cancer cell. This mechanism enabled the development of peptide-based vaccines that can activate the patient's immune response to treat cancers. Hence, the ability of accurately predict peptide-MHC binding is an essential component for prioritizing the best peptides for each patient. However, peptide-MHC binding experimental data for many MHC alleles are still lacking, which limited the accuracy of existing prediction models. RESULTS: In this study, we presented an improved version of MHCSeqNet that utilized sub-word-level peptide features, a 3D structure embedding for MHC alleles, and an expanded training dataset to achieve better generalizability on MHC alleles with small amounts of data. Visualization of MHC allele embeddings confirms that the model was able to group alleles with similar binding specificity, including those with no peptide ligand in the training dataset. Furthermore, an external evaluation suggests that MHCSeqNet2 can improve the prioritization of T cell epitopes for MHC alleles with small amount of training data. AVAILABILITY AND IMPLEMENTATION: The source code and installation instruction for MHCSeqNet2 are available at https://github.com/cmb-chula/MHCSeqNet2.
Subject(s)
Histocompatibility Antigens Class I , Peptides , Humans , Alleles , Histocompatibility Antigens Class I/genetics , Histocompatibility Antigens Class I/chemistry , Peptides/chemistry , Protein Binding , Epitopes, T-Lymphocyte/metabolismABSTRACT
The emergence of Omicron as the fifth variant of concern within the SARS-CoV-2 pandemic in late 2021, characterized by its rapid transmission and distinct spike gene mutations, underscored the pressing need for cost-effective and efficient methods to detect viral variants, especially given their evolving nature. This study sought to address this need by assessing the effectiveness of two SARS-CoV-2 variant classification platforms based on RT-PCR and mass spectrometry. The primary aim was to differentiate between Delta, Omicron BA.1, and Omicron BA.2 variants using 618 COVID-19-positive samples collected from Bangkok patients between November 2011 and March 2022. The analysis revealed that both BA.1 and BA.2 variants exhibited significantly higher transmission rates, up to 2-3 times, when compared to the Delta variant. This research presents a cost-efficient approach to virus surveillance, enabling a quantitative evaluation of variant-specific public health implications, crucial for informing and adapting public health strategies.
ABSTRACT
Neurofilament light chain has become a promising biomarker for neuroaxonal injury; however, its diagnostic utility is limited to chronic disorders or specific contexts. Alteration of consciousness is a common clinical problem with diverse aetiologies, many of which require timely diagnoses. We evaluated the value of neurofilament light chain alone, as well as creating diagnostic models, in distinguishing causes of alteration of consciousness. Patients presenting with alteration of consciousness were enrolled. Initial clinical data of each participant were evaluated by a neurologist to give a provisional diagnosis. Each participant subsequently received advanced investigations and follow-up to conclude the final diagnosis. All diagnoses were classified into a structural or non-structural cause of alteration of consciousness. Plasma and cerebrospinal fluid levels of neurofilament light chain were measured. Cerebrospinal fluid neurofilament light chain and other clinical parameters were used to develop logistic regression models. The performance of cerebrospinal fluid neurofilament light chain, the neurologist's provisional diagnosis, and the model to predict the final diagnosis were compared. For the results, among 71 participants enrolled, 67.6% and 32.4% of their final diagnoses were classified as structural and non-structural, respectively. Cerebrospinal fluid neurofilament light chain demonstrated an area under the curve of 0.75 (95% confidence interval 0.63-0.88) which was not significantly different from a neurologist's provisional diagnosis 0.85 (95% confidence interval 0.75-0.94) (P = 0.14). The multivariable regression model using cerebrospinal fluid neurofilament light chain and other basic clinical data achieved an area under the curve of 0.90 (95% confidence interval 0.83-0.98). In conclusion, neurofilament light chain classified causes of alteration of consciousness with moderate accuracy. Nevertheless, including other basic clinical data to construct a model improved the performance to a level that was comparable to clinical neurologists.
ABSTRACT
Next generation sequencing of circulating tumor DNA (ctDNA) has been used as a noninvasive alternative for cancer diagnosis and characterization of tumor mutational landscape. However, low ctDNA fraction and other factors can limit the ability of ctDNA analysis to capture tumor-specific and actionable variants. In this study, whole-exome sequencings (WES) were performed on paired ctDNA and tumor biopsy in 15 cancer patients to assess the extent of concordance between mutational profiles derived from the two source materials. We found that up to 16.4% ctDNA fraction can still be insufficient for detecting tumor-specific variants and that good concordance with tumor biopsy is consistently achieved at higher ctDNA fractions. Most importantly, ctDNA analysis can consistently capture tumor heterogeneity and detect key cancer-related genes even in a patient with both primary and metastatic tumors.
Subject(s)
Circulating Tumor DNA , Neoplasms , Humans , Circulating Tumor DNA/genetics , Exome Sequencing , Biomarkers, Tumor/genetics , Neoplasms/diagnosis , Neoplasms/genetics , Neoplasms/pathology , Mutation , High-Throughput Nucleotide SequencingABSTRACT
When planning radiation therapy, late effects due to the treatment should be considered. One of the most common complications of head and neck radiation therapy is hypothyroidism. Although clinical and dosimetric data are routinely used to assess the risk of hypothyroidism after radiation, the outcome is still unsatisfactory. Medical imaging can provide additional information that improves the prediction of hypothyroidism. In this study, pre-treatment computed tomography (CT) radiomics features of the thyroid gland were combined with clinical and dosimetric data from 220 participants to predict the occurrence of hypothyroidism within 2 years after radiation therapy. The findings demonstrated that the addition of CT radiomics consistently and significantly improves upon conventional model, achieving the highest area under the receiver operating characteristic curve (AUCs) of 0.81 ± 0.06 with a random forest model. Hence, pre-treatment thyroid CT imaging provides useful information that have the potential to improve the ability to predict hypothyroidism after nasopharyngeal radiation therapy.
Subject(s)
Hypothyroidism , Nasopharyngeal Neoplasms , Humans , Nasopharyngeal Carcinoma/radiotherapy , Nasopharyngeal Carcinoma/complications , Hypothyroidism/diagnostic imaging , Hypothyroidism/etiology , Hypothyroidism/epidemiology , Tomography, X-Ray Computed/methods , Nasopharyngeal Neoplasms/diagnostic imaging , Nasopharyngeal Neoplasms/radiotherapy , Nasopharyngeal Neoplasms/complications , Retrospective StudiesABSTRACT
Human neutrophil peptides (HNPs) can induce cell proliferation and activation so their growth promoting activities may have potential clinical benefit. This study investigated the effects of HNPs on human dermal fibroblasts. Differential gene expression in HNP-treated cells and genes involved in regulating intracellular pathways were explored. Dermal fibroblasts were isolated from healthy neonatal foreskin and treated with HNPs in 2D and 3D cell culture systems. The expression of cell proliferation (Ki-67) gene and cell activation (COL1A1) gene plus their proteins was measured. Differential gene expression was determined using RNA-seq, and upregulated and downregulated genes were mapped onto intracellular pathways by KEGG analysis and Gene Ontology databases. HNPs significantly increased cell proliferation without cytotoxicity whilst HNP1 enhanced expression of COL1A1 and type I collagen production in 2D cells and 3D spheroids. RNA-sequencing analysis showed gene clustering with clear separation between HNP1-treated and control groups. A heatmap of top 50 differentially expressed genes was consistent among HNP1-treated samples. Most upregulated genes were associated with cell proliferation and activation as mapped into intracellular pathways whilst most downregulated genes belonged to steroid/arachidonic acid metabolism and inflammatory signaling pathways. HNP1 increased cell proliferation and activation but reduced lipid metabolism and inflammation.
Subject(s)
Neutrophils , alpha-Defensins , Infant, Newborn , Humans , Neutrophils/metabolism , alpha-Defensins/metabolism , Signal Transduction , Skin/metabolism , Fibroblasts/metabolismABSTRACT
Many studies have shown that cellular morphology can be used to distinguish spiked-in tumor cells in blood sample background. However, most validation experiments included only homogeneous cell lines and inadequately captured the broad morphological heterogeneity of cancer cells. Furthermore, normal, non-blood cells could be erroneously classified as cancer because their morphology differ from blood cells. Here, we constructed a dataset of microscopic images of organoid-derived cancer and normal cell with diverse morphology and developed a proof-of-concept deep learning model that can distinguish cancer cells from normal cells within an unlabeled microscopy image. In total, more than 75,000 organoid-drived cells from 3 cholangiocarcinoma patients were collected. The model achieved an area under the receiver operating characteristics curve (AUROC) of 0.78 and can generalize to cell images from an unseen patient. These resources serve as a foundation for an automated, robust platform for circulating tumor cell detection.
Subject(s)
Cell Line, Tumor , Neoplasms , Humans , Area Under Curve , Deep Learning , Microscopy , Cell Line, Tumor/classification , Cell Line, Tumor/pathology , Neoplasms/diagnostic imaging , Neoplasms/pathologyABSTRACT
BACKGROUND: The peroxisome is a ubiquitous single membrane-enclosed organelle with an important metabolic role. Peroxisomal disorders represent a class of medical conditions caused by deficiencies in peroxisome function and are segmented into enzyme-and-transporter defects (defects in single peroxisomal proteins) and peroxisome biogenesis disorders (defects in the peroxin proteins, critical for normal peroxisome assembly and biogenesis). In this study, we employed multivariate supervised and non-supervised statistical methods and utilized mass spectrometry data of neurological patients, peroxisomal disorder patients (X-linked adrenoleukodystrophy and Zellweger syndrome), and healthy controls to analyze the role of common metabolites in peroxisomal disorders, to develop and refine a classification models of X-linked adrenoleukodystrophy and Zellweger syndrome, and to explore analytes with utility in rapid screening and diagnostics. RESULTS: T-SNE, PCA, and (sparse) PLS-DA, operated on mass spectrometry data of patients and healthy controls were utilized in this study. The performance of exploratory PLS-DA models was assessed to determine a suitable number of latent components and variables to retain for sparse PLS-DA models. Reduced-features (sparse) PLS-DA models achieved excellent classification performance of X-linked adrenoleukodystrophy and Zellweger syndrome patients. CONCLUSIONS: Our study demonstrated metabolic differences between healthy controls, neurological patients, and peroxisomal disorder (X-linked adrenoleukodystrophy and Zellweger syndrome) patients, refined classification models and showed the potential utility of hexacosanoylcarnitine (C26:0-carnitine) as a screening analyte for Chinese patients in the context of a multivariate discriminant model predictive of peroxisomal disorders.
Subject(s)
Adrenoleukodystrophy , Peroxisomal Disorders , Zellweger Syndrome , Child , Humans , Adrenoleukodystrophy/diagnosis , East Asian People , Multivariate Analysis , Peroxisomal Disorders/diagnosis , Peroxisomal Disorders/metabolism , Zellweger Syndrome/diagnosis , Zellweger Syndrome/metabolism , ChinaABSTRACT
Mitotic count (MC) is an important histological parameter for cancer diagnosis and grading, but the manual process for obtaining MC from whole-slide histopathological images is very time-consuming and prone to error. Therefore, deep learning models have been proposed to facilitate this process. Existing approaches utilize a two-stage pipeline: the detection stage for identifying the locations of potential mitotic cells and the classification stage for refining prediction confidences. However, this pipeline formulation can lead to inconsistencies in the classification stage due to the poor prediction quality of the detection stage and the mismatches in training data distributions between the two stages. In this study, we propose a Refine Cascade Network (ReCasNet), an enhanced deep learning pipeline that mitigates the aforementioned problems with three improvements. First, window relocation was used to reduce the number of poor quality false positives generated during the detection stage. Second, object re-cropping was performed with another deep learning model to adjust poorly centered objects. Third, improved data selection strategies were introduced during the classification stage to reduce the mismatches in training data distributions. ReCasNet was evaluated on two large-scale mitotic figure recognition datasets, canine cutaneous mast cell tumor (CCMCT) and canine mammary carcinoma (CMC), which resulted in up to 4.8% percentage point improvements in the F1 scores for mitotic cell detection and 44.1% reductions in mean absolute percentage error (MAPE) for MC prediction. Techniques that underlie ReCasNet can be generalized to other two-stage object detection pipeline and should contribute to improving the performances of deep learning models in broad digital pathology applications.
Subject(s)
Mitosis , Animals , DogsABSTRACT
Since its establishment in 2015, the transcriptomics-based consensus molecular subtype (CMS) classification has unified our understanding of colorectal cancer. Each of the four CMS exhibited distinctive high-level molecular signatures that correlated well with prognosis and treatment response. Nonetheless, many key aspects of colorectal cancer progression and intra-subtype heterogeneity remain unresolved. This is partly because the bulk transcriptomic data used to define CMS contain substantial interference from non-tumor cells. Here, we propose a concise panel of 62 genes that not only accurately recapitulates all key characteristics of the four original CMS but also identifies three additional subpopulations with unique molecular signatures. Validation on independent cohorts confirms that the new CMS4 intra-subtypes coincide with single-cell-derived intrinsic subtypes and that the panel consists of many immune cell-type markers that can capture the status of tumor microenvironment. Furthermore, a 2D embedding of CMS structure based on the proposed gene panel provides a high-resolution view of the functional pathways and cell-type markers that underlie each CMS intra-subtype and the continuous progression from CMS2 to CMS4 subtypes. Our gene panel and 2D visualization refined the delineation of colorectal cancer subtypes and could aid further discovery of molecular mechanisms in colorectal cancer. IMPLICATIONS: : Well-selected gene panel and representation can capture both the continuum of cancer cell states and tumor microenvironment status.
Subject(s)
Colorectal Neoplasms , Humans , Colorectal Neoplasms/pathology , Gene Expression Profiling , Transcriptome , Biomarkers, Tumor/genetics , Tumor Microenvironment/geneticsABSTRACT
PURPOSE: The aim of this study was to develop a normal tissue complication probability model using a machine learning approach (ML-based NTCP) to predict the risk of radiation-induced liver disease in hepatocellular carcinoma (HCC) patients. MATERIALS AND METHODS: The study population included 201 HCC patients treated with radiotherapy. The patients' medical records were retrospectively reviewed to obtain the clinical and radiotherapy data. Toxicity was defined by albumin-bilirubin (ALBI) grade increase. The normal liver dose-volume histogram was reduced to mean liver dose (MLD) based on the fraction size-adjusted equivalent uniform dose (2 Gy/fraction and α/ß = 2). Three types of ML-based classification models were used, a penalized logistic regression (PLR), random forest (RF), and gradient-boosted tree (GBT) model. Model performance was compared using the area under the receiver operating characteristic curve (AUROC). Internal validation was performed by 5-fold cross validation and external validation was done in 44 new patients. RESULTS: Liver toxicity occurred in 87 patients (43.1%). The best individual model was the GBT model using baseline liver function, liver volume, and MLD as inputs and the best overall model was an ensemble of the PLR and GBT models. An AUROC of 0.82 with a standard deviation of 0.06 was achieved for the internal validation. An AUROC of 0.78 with a standard deviation of 0.03 was achieved for the external validation. The behaviors of the best GBT model were also in good agreement with the domain knowledge on NTCP. CONCLUSION: We propose the methodology to develop an ML-based NTCP model to estimate the risk of ALBI grade increase.