Search PubMed⌕ Search

SEARCH · Search PubMed

Results for “Supervised Machine Learning”

Search indexed PubMed citations on genomics, clinical trials, systematic reviews and public health. Explore titles, authors and supplied subject terms, then open the PubMed record.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 37 records · Page 2Linked to original sources
PubMed2025

Alzheimer's subtypes A supervised, unsupervised, multimodal, multilayered embedded recursive (SUMMER) AI study.

Since Alzheimer's disease (AD) is a heterogeneous disease, different subtypes may have distinct biological, genetic, and clinical characteristics, requiring tailored interventions. While several proposed subtypes of AD exist, there is still no clear consensus on a definitive classification. By leveraging complementary AI approaches, including supervised and unsupervised learning, within a recursive pipeline (SUMMER) that integrates multimodal datasets encompassing MRI measurements, phenotypes, and genetic data, our goal was to generate robust scientific evidence for identifying AD subtypes. Data was downloaded from the Alzheimer's Disease Neuroimaging Initiative (ADNI) database and included neuroimaging data (MRI), genetics (SNPs), clinical diagnosis, and demographics. 1133 European American participants' images, aged 55-95, were included in this study. The analysis was multi-fold, where the first step involved applying an unsupervised application to a subset of the MRI sample (AD + cognitively normal (CN) aged matched groups, 100 men aged 68-85 years, and 76 women aged 68-85 years). The MRI brain gray matter was segmented into 44 regions of interest (ROIs) according to a standard atlas, and 618 features were extracted, including ROI voxel intensity measurements such as minimum, maximum, and histogram variables. Results identified a cluster of subtype AD men and a cluster of subtype AD women that were distinct from the rest of their respective samples. In the next step, the integrity of the identified subtype AD clusters was investigated using the XGBoost supervised machine learning application with genetic features (SNPs, N=36,724) and labels: the identified subtype AD cluster vs. the rest of the sample, stratified by sex. A significant AD subtype men model (accuracy=0.85, F1=0.72, AUC=0.83) and a significant women AD subtype model (accuracy=0.81, F1=0.81, AUC=0.81) were built, confirming the homogeneity of the isolated AD subtype clusters. Discriminative biomarkers were extracted from the significant models, including selected ROIs and SNPs. Finally, the subtype models were tested on an unseen subset of ADNI data. The genetic-based models identified clusters of AD subtype participants consisting of 34% of the men AD group and 47% of the women AD group. Phenotypic analysis indicates that lower body weight was associated with the women's AD subtype. Complex diseases like AD demand a sophisticated, multimodal approach for precise diagnosis. Effectively identifying disease subtypes enhances the potential for personalized treatment, ultimately improving patient outcomes.

Journal Article↗
PubMed2004

Automated interpretation of subcellular patterns from immunofluorescence microscopy.

Immunofluorescence microscopy is widely used to analyze the subcellular locations of proteins, but current approaches rely on visual interpretation of the resulting patterns. To facilitate more rapid, objective, and sensitive analysis, computer programs have been developed that can identify and compare protein subcellular locations from fluorescence microscope images. The basis of these programs is a set of features that numerically describe the characteristics of protein images. Supervised machine learning methods can be used to learn from the features of training images and make predictions of protein location for images not used for training. Using image databases covering all major organelles in HeLa cells, these programs can achieve over 92% accuracy for two-dimensional (2D) images and over 95% for three-dimensional images. Importantly, the programs can discriminate proteins that could not be distinguished by visual examination. In addition, the features can also be used to rigorously compare two sets of images (e.g., images of a protein in the presence and absence of a drug) and to automatically select the most typical image from a set. The programs described provide an important set of tools for those using fluorescence microscopy to study protein location.

Automation↗
PubMed2002

Identifying genes related to drug anticancer mechanisms using support vector machine.

In an effort to identify genes related to the cell line chemosensitivity and to evaluate the functional relationships between genes and anticancer drugs acting by the same mechanism, a supervised machine learning approach called support vector machine was used to label genes into any of the five predefined anticancer drug mechanistic categories. Among dozens of unequivocally categorized genes, many were known to be causally related to the drug mechanisms. For example, a few genes were found to be involved in the biological process triggered by the drugs (e.g. DNA polymerase epsilon was the direct target for the drugs from DNA antimetabolites category). DNA repair-related genes were found to be enriched for about eight-fold in the resulting gene set relative to the entire gene set. Some uncharacterized transcripts might be of interest in future studies. This method of correlating the drugs and genes provides a strategy for finding novel biologically significant relationships for molecular pharmacology.

Antineoplastic Agents↗
PubMed2001

Feature subset selection by genetic algorithms and estimation of distribution algorithms. A case study in the survival of cirrhotic patients treated with TIPS.

The transjugular intrahepatic portosystemic shunt (TIPS) is an interventional treatment for cirrhotic patients with portal hypertension. In the light of our medical staff's experience, the consequences of TIPS are not homogeneous for all the patients and a subgroup dies in the first 6 months after TIPS placement. Actually, there is no risk indicator to identify this subgroup of patients before treatment. An investigation for predicting the survival of cirrhotic patients treated with TIPS is carried out using a clinical database with 107 cases and 77 attributes. Four supervised machine learning classifiers are applied to discriminate between both subgroups of patients. The application of several feature subset selection (FSS) techniques has significantly improved the predictive accuracy of these classifiers and considerably reduced the amount of attributes in the classification models. Among FSS techniques, FSS-TREE, a new randomized algorithm inspired on the new EDA (estimation of distribution algorithm) paradigm has obtained the best average accuracy results for each classifier.

Algorithms↗
PubMed2003

Analyzing tumor gene expression profiles.

A brief introduction to high throughput technologies for measuring and analyzing gene expression is given. Various supervised and unsupervised data mining methods for analyzing the produced high-dimensional data are discussed. The main emphasis is on supervised machine learning methods for classification and prediction of tumor gene expression profiles. Furthermore, methods to rank the genes according to their importance for the classification are explored. The approaches are illustrated by exploratory studies using two examples of retrospective clinical data from routine tests; diagnostic prediction of small round blue cell tumors (SRBCT) of childhood and determining the estrogen receptor (ER) status of sporadic breast cancer. The classification performance is gauged using blind tests. These studies demonstrate the feasibility of machine learning-based molecular cancer classification.

Adult↗
PubMed2005

Controlled protein precipitation in combination with chip-based nanospray infusion mass spectrometry. An approach for metabolomics profiling of plasma.

Liquid chromatography-mass spectrometry (LC-MS) is a common method for profiling biological samples in metabolomics. However, LC-MS data of metabolomic studies are often affected by high noise levels, retention time shifts, and high variability in signal intensities. With a new chip-based nanoelectrospray source it becomes possible to directly infuse complex biological samples such as plasma without any chromatographic separation beforehand. In combination with highly diluted samples and long data acquisition times, the parallel analysis of hundreds of compounds is now possible. In a proof-of-concept study, 10 human plasma samples from females and males were analyzed with the intention to separate the two groups by their different metabolomes. The reproducibility was so high that statistical analysis of the data could be performed without prior normalization. Two groups of female and male samples were separated by a supervised machine learning algorithm, principal component analysis, and hierarchical clustering. Peaks contributing to the group separation were characterized by accurate mass measurement and MS-MS fragmentation and by spiking experiments. The feasibility of direct sample infusion using the new chip-based nanoelectrospray device opens a new dimension for the rapid parallel analysis of complex biological mixtures.

Adult↗
PubMedDate not supplied

Data mining the NCI cancer cell line compound GI(50) values: identifying quinone subtypes effective against melanoma and leukemia cell classes.

Using data mining techniques, we have studied a subset (1400) of compounds from the large public National Cancer Institute (NCI) compounds data repository. We first carried out a functional class identity assignment for the 60 NCI cancer testing cell lines via hierarchical clustering of gene expression data. Comprised of nine clinical tissue types, the 60 cell lines were placed into six classes-melanoma, leukemia, renal, lung, and colorectal, and the sixth class was comprised of mixed tissue cell lines not found in any of the other five classes. We then carried out supervised machine learning, using the GI(50) values tested on a panel of 60 NCI cancer cell lines. For separate 3-class and 2-class problem clustering, we successfully carried out clear cell line class separation at high stringency, p < 0.01 (Bonferroni corrected t-statistic), using feature reduction clustering algorithms embedded in RadViz, an integrated high dimensional analytic and visualization tool. We started with the 1400 compound GI(50) values as input and selected only those compounds, or features, significant in carrying out the classification. With this approach, we identified two small sets of compounds that were most effective in carrying out complete class separation of the melanoma, non-melanoma classes and leukemia, non-leukemia classes. To validate these results, we showed that these two compound sets' GI(50) values were highly accurate classifiers using five standard analytical algorithms. One compound set was most effective against the melanoma class cell lines (14 compounds), and the other set was most effective against the leukemia class cell lines (30 compounds). The two compound classes were both significantly enriched in two different types of substituted p-quinones. The melanoma cell line class of 14 compounds was comprised of 11 compounds that were internal substituted p-quinones, and the leukemia cell line class of 30 compounds was comprised of 6 compounds that were external substituted p-quinones. Attempts to subclassify melanoma or leukemia cell lines based upon their clinical cancer subtype met with limited success. For example, using GI(50) values for the 30 compounds we identified as effective against all leukemia cell lines, we could subclassify acute lymphoblastic leukemia (ALL) origin cell lines from non-ALL leukemia origin cell lines without significant overlap from non-leukemia cell lines. Based upon clustering using GI(50) values for the 60 cancer cell lines laid out by the RadViz algorithm, these two compound subsets did not overlap with clusters containing any of the NCI's 92 compounds of known mechanism of action, a few of which are quinones. Given their structural patterns, the two p-quinone subtypes we identified would clearly be expected to possess different redox potentials/substrate specificities for enzymatic reduction in vivo. These two p-quinone subtypes represent valuable information that may be used in the elucidation of pharmacophores for the design of compounds to treat these two cancer tissue types in the clinic.

Algorithms↗
PubMed2026

Social disconnection integrates genetic and proteomic risks in suicidal ideation and depression.

Suicidal ideation (SI) and major depressive disorder (MDD) are complex psychiatric conditions arising from the interplay of genetic liability, molecular processes, and psychosocial factors. While these dimensions have been extensively studied in isolation, their joint contribution to SI and MDD remains unclear. This study integrates multi-modal data to elucidate these synergistic effects and develop robust models for individual-level risk stratification. Leveraging longitudinal multi-modal data from 13,085 UK Biobank participants, we integrated genomic, proteomic, and social connection profiles. We developed interpretable risk scores using a rigorous supervised machine learning framework encompassing diverse linear and ensemble classifiers. Permutation importance was employed to quantify feature contributions and derive transparent, weighted risk metrics across diverse classifiers. These scores were validated through association, interaction, and mediation analyses. Social connection-based risk scores significantly differentiated cases and controls across the two suicidal ideation phenotypes at 2017 and 2023 with cross-sectional analyses (AUCs: 0.70 - 0.73), outperforming proteomic-only models. Functional dimensions of social connection emerged as the most informative predictors. Longitudinal analyses revealed that social risk scores at baseline predicted suicidal ideation onset six years later, independent of demographic covariates. Interaction analyses demonstrated that polygenic risk for suicide attempt significantly interacted with both social and proteomic risk features in relation to depression. Structural equation models further confirmed that social disconnection acts as a key mediator linking genetic predisposition to MDD and SI. Social disconnection is a critical risk factor mediating the impact of genetic vulnerability on psychiatric outcomes. Integrating social, genetic, and molecular data supports a multilevel framework for risk stratification and highlights the potential of socially oriented interventions to mitigate biological risk.

Humans↗
PubMed2026

An encyclopedia of human enhancer-gene regulatory interactions.

Identifying transcriptional enhancers and their target genes is essential for understanding gene regulation and the effect of human genetic variation on disease1-6. Here we create and evaluate a resource of more than 92&#x2009;million enhancer-gene regulatory interactions across 1,458 biosamples covering 369 cell types and tissues, by integrating predictive models, chromatin states, three-dimensional contacts and large-scale genetic perturbations generated by the ENCODE Consortium7. We first create a systematic benchmarking pipeline to compare predictive models, assembling a dataset of 10,356 element-gene pairs measured in CRISPR perturbation experiments, more than 30,000 fine-mapped expression quantitative trait loci and 569 fine-mapped genome-wide association study&#xa0;(GWAS) variants linked to a probable causal gene. Using this framework, we develop ENCODE-rE2G, a predictive model achieving state-of-the-art performance across several prediction tasks, demonstrating that iterative perturbations and supervised machine learning can build increasingly accurate predictive models of enhancer regulation. Using ENCODE-rE2G, we build an encyclopedia of enhancer-gene regulatory interactions in the human genome, revealing global properties of enhancer networks, identifying differences in regulatory complexity across genes and improving analyses linking noncoding variants to target genes and cell types for common complex diseases. By interpreting the model, we find that beyond enhancer activity and three-dimensional enhancer-promoter contacts, additional features that&#xa0;guide enhancer-promoter communication include promoter class and enhancer-enhancer synergy. These genome-wide maps of enhancer-gene regulatory interactions, benchmarking software, predictive models and insights about enhancer function provide a valuable resource for future studies of gene regulation and human genetics.

Humans↗
PubMed2004

Mismatch string kernels for discriminative protein classification.

MOTIVATION: Classification of proteins sequences into functional and structural families based on sequence homology is a central problem in computational biology. Discriminative supervised machine learning approaches provide good performance, but simplicity and computational efficiency of training and prediction are also important concerns. RESULTS: We introduce a class of string kernels, called mismatch kernels, for use with support vector machines (SVMs) in a discriminative approach to the problem of protein classification and remote homology detection. These kernels measure sequence similarity based on shared occurrences of fixed-length patterns in the data, allowing for mutations between patterns. Thus, the kernels provide a biologically well-motivated way to compare protein sequences without relying on family-based generative models such as hidden Markov models. We compute the kernels efficiently using a mismatch tree data structure, allowing us to calculate the contributions of all patterns occurring in the data in one pass while traversing the tree. When used with an SVM, the kernels enable fast prediction on test sequences. We report experiments on two benchmark SCOP datasets, where we show that the mismatch kernel used with an SVM classifier performs competitively with state-of-the-art methods for homology detection, particularly when very few training examples are available. Examination of the highest-weighted patterns learned by the SVM classifier recovers biologically important motifs in protein families and superfamilies.

Algorithms↗
PubMed2004

Extracting and characterizing gene-drug relationships from the literature.

A fundamental task of pharmacogenetics is to collect and classify relationships between genes and drugs. Currently, this useful information has not been comprehensively aggregated in any database and remains scattered throughout the published literature. Although there are efforts to collect this information manually, they are limited by the size of the published literature on gene-drug relationships. Therefore, we investigated computational methods to extract and characterize pharmacogenetic relationships between genes and drugs from the literature. We first evaluated the effectiveness of the co-occurrence method in identifying related genes and drugs. We then used supervised machine learning algorithms to classify the relationships between genes and drugs from the Pharmacogenetics and Pharmacogenomics Knowledge Base (PharmGKB) into five categories that have been defined by active pharmacogenetic researchers as relevant to their work. The final co-occurrence algorithm was able to extract 78% of the related genes and drugs that were published in a review article from the literature. Our algorithm subsequently classified the relationships between genes and drugs from the PharmGKB into five categories with 74% accuracy. We have made the data available on a supplementary website at http://bionlp.stanford.edu/genedrug/ Gene-drug relationships can be accurately extracted from text and classified into categories. Although the relationships that we have identified do not capture the details and fine distinctions often made in the literature, these methods will help scientists to track the ever-growing literature and create information resources to support future discoveries.

Algorithms↗
PubMed2007

Two-sample comparison based on prediction error, with applications to candidate gene association studies.

To take advantage of the increasingly available high-density SNP maps across the genome, various tests that compare multilocus genotypes or estimated haplotypes between cases and controls have been developed for candidate gene association studies. Here we view this two-sample testing problem from the perspective of supervised machine learning and propose a new association test. The approach adopts the flexible and easy-to-understand classification tree model as the learning machine, and uses the estimated prediction error of the resulting prediction rule as the test statistic. This procedure not only provides an association test but also generates a prediction rule that can be useful in understanding the mechanisms underlying complex disease. Under the set-up of a haplotype-based transmission/disequilibrium test (TDT) type of analysis, we find through simulation studies that the proposed procedure has the correct type I error rates and is robust to population stratification. The power of the proposed procedure is sensitive to the chosen prediction error estimator. Among commonly used prediction error estimators, the .632+ estimator results in a test that has the best overall performance. We also find that the test using the .632+ estimator is more powerful than the standard single-point TDT analysis, the Pearson's goodness-of-fit test based on estimated haplotype frequencies, and two haplotype-based global tests implemented in the genetic analysis package FBAT. To illustrate the application of the proposed method in population-based association studies, we use the procedure to study the association between non-Hodgkin lymphoma and the IL10 gene.

Adult↗
PubMed2005

Multiclass molecular cancer classification by kernel subspace methods with effective kernel parameter selection.

Microarray techniques provide new insights into molecular classification of cancer types, which is critical for cancer treatments and diagnosis. Recently, an increasing number of supervised machine learning methods have been applied to cancer classification problems using gene expression data. Support vector machines (SVMs), in particular, have become one of the most effective and leading methods. However, there exist few studies on the application of other kernel methods in the literature. We apply a kernel subspace (KS) method to multiclass cancer classification problems, and assess its validity by comparing it with multiclass SVMs. Our comparative study using seven multiclass cancer datasets demonstrates that the KS method has high performance that is comparable to multiclass SVMs. Furthermore, we propose an effective criterion for kernel parameter selection, which is shown to be useful for the computation of the KS method.

Artificial Intelligence↗
PubMed2004

Statistical evaluation of local alignment features predicting allergenicity using supervised classification algorithms.

BACKGROUND: Recently, two promising alignment-based features predicting food allergenicity using the k nearest neighbor (kNN) classifier were reported. These features are the alignment score and alignment length of the best local alignment obtained in a database of known allergen sequences. METHODS: In the work reported here a much more comprehensive statistical evaluation of the potential of these features was performed, this time for the prediction of allergenicity in general. The evaluation consisted of the following four key components. (1) A new high quality database consisting of 318 carefully selected, non-redundant allergens and 1,007 sequences carefully selected to be non-allergens. (2) Three different supervised algorithms: the kNN classifier, the Bayesian linear Gaussian classifier, and the Bayesian quadratic Gaussian classifier. (3) A large set of local alignment procedures defined using the FASTA3 alignment program by means of a wide range of different parameter settings. (4) Novel performance curves, alternative to conventional receiver-operating characteristic curves, to display not only average behaviors but also statistical variations due to small data sets. RESULTS: The linear Gaussian classifier proved most useful among the tested supervised machine learning algorithms, closely followed by the quadratic Gaussian equivalent and kNN. The overall best classification results were obtained with a novel feature vector consisting of the combined alignment scores derived from local alignment procedures using different substitution matrices. CONCLUSIONS: The models reported here should be useful as a part of an integrated assessment scheme for potential protein allergenicity and for future comparisons with alternative bioinformatic approaches.

Algorithms↗
PubMed2026

A validated, modifiable proteomic score from the EXSCEL trial predicts cardiovascular events in diabetes.

BACKGROUNDAdults with type 2 diabetes mellitus (T2DM) are at increased risk for stroke, myocardial infarction, and cardiovascular death, yet individual risk is heterogeneous and incompletely captured by clinical models.METHODSIn the Exenatide Study of Cardiovascular Event Lowering (EXSCEL), adults with T2DM were randomized to a GLP-1 RA (exenatide) or a placebo and followed longitudinally for major adverse cardiovascular events (MACE). High-throughput discovery proteomics was done in plasma collected at baseline and 12 months. Proteins associated with time to MACE were identified using multivariable regression and incorporated into supervised machine learning models. A multi-protein score was developed and externally validated in 2 independent population-based and trial cohorts.RESULTSThe proteomic score showed incremental improvement in cardiovascular risk discrimination beyond clinical factors alone, and several proteins were consistently prioritized across modeling approaches. The protein score and a top-ranked protein, tetranectin, were modified by GLP-1 RA treatment, and a decrease in protein score was associated with improved outcomes, supporting modifiability of MACE risk.CONCLUSIONExternal validation confirmed generalizability across cohorts with and without diabetes. Together, these findings demonstrate that plasma proteomic signatures can enhance cardiovascular risk stratification and identify treatment-responsive biomarkers in T2DM, supporting their potential role in precision prevention strategiesFUNDINGThe EXSCEL study was funded by Amylin Pharmaceuticals. This research was supported by contracts HHSN268201200036C, HHSN268200800007C, HHSN268201800001C, N01HC55222, N01HC85079, N01HC85080, N01HC85081, N01HC85082, N01HC85083, N01HC85086, 75N92021D00006, and grants R01HL146145, U01HL080295, U01HL130114, R01HL172803, and R01HL144483 from the National Heart, Lung, and Blood Institute, with additional contribution from the National Institute of Neurological Disorders and Stroke. Additional support was provided by R01AG023629 from the National Institute on Aging.

Aged↗
PubMed2004

Predicting co-complexed protein pairs using genomic and proteomic data integration.

BACKGROUND: Identifying all protein-protein interactions in an organism is a major objective of proteomics. A related goal is to know which protein pairs are present in the same protein complex. High-throughput methods such as yeast two-hybrid (Y2H) and affinity purification coupled with mass spectrometry (APMS) have been used to detect interacting proteins on a genomic scale. However, both Y2H and APMS methods have substantial false-positive rates. Aside from high-throughput interaction screens, other gene- or protein-pair characteristics may also be informative of physical interaction. Therefore it is desirable to integrate multiple datasets and utilize their different predictive value for more accurate prediction of co-complexed relationship. RESULTS: Using a supervised machine learning approach--probabilistic decision tree, we integrated high-throughput protein interaction datasets and other gene- and protein-pair characteristics to predict co-complexed pairs (CCP) of proteins. Our predictions proved more sensitive and specific than predictions based on Y2H or APMS methods alone or in combination. Among the top predictions not annotated as CCPs in our reference set (obtained from the MIPS complex catalogue), a significant fraction was found to physically interact according to a separate database (YPD, Yeast Proteome Database), and the remaining predictions may potentially represent unknown CCPs. CONCLUSIONS: We demonstrated that the probabilistic decision tree approach can be successfully used to predict co-complexed protein (CCP) pairs from other characteristics. Our top-scoring CCP predictions provide testable hypotheses for experimental validation.

Computational Biology↗
PubMed2006

Prediction of catalytic residues using Support Vector Machine with selected protein sequence and structural properties.

BACKGROUND: The number of protein sequences deriving from genome sequencing projects is outpacing our knowledge about the function of these proteins. With the gap between experimentally characterized and uncharacterized proteins continuing to widen, it is necessary to develop new computational methods and tools for functional prediction. Knowledge of catalytic sites provides a valuable insight into protein function. Although many computational methods have been developed to predict catalytic residues and active sites, their accuracy remains low, with a significant number of false positives. In this paper, we present a novel method for the prediction of catalytic sites, using a carefully selected, supervised machine learning algorithm coupled with an optimal discriminative set of protein sequence conservation and structural properties. RESULTS: To determine the best machine learning algorithm, 26 classifiers in the WEKA software package were compared using a benchmarking dataset of 79 enzymes with 254 catalytic residues in a 10-fold cross-validation analysis. Each residue of the dataset was represented by a set of 24 residue properties previously shown to be of functional relevance, as well as a label {+1/-1} to indicate catalytic/non-catalytic residue. The best-performing algorithm was the Sequential Minimal Optimization (SMO) algorithm, which is a Support Vector Machine (SVM). The Wrapper Subset Selection algorithm further selected seven of the 24 attributes as an optimal subset of residue properties, with sequence conservation, catalytic propensities of amino acids, and relative position on protein surface being the most important features. CONCLUSION: The SMO algorithm with 7 selected attributes correctly predicted 228 of the 254 catalytic residues, with an overall predictive accuracy of more than 86%. Missing only 10.2% of the catalytic residues, the method captures the fundamental features of catalytic residues and can be used as a "catalytic residue filter" to facilitate experimental identification of catalytic residues for proteins with known structure but unknown function.

Algorithms↗
PubMed2025

Tumor-immune partitioning and clustering algorithm for identifying tumor-immune cell spatial interaction signatures within the tumor microenvironment.

BACKGROUND: Growing evidence supports the importance of characterizing the organizational patterns of various cellular constituents in the tumor microenvironment in precision oncology. Most existing data on immune cell infiltrates in tumors, which are based on immune cell counts or nearest neighbor-type analyses, have failed to fully capture the cellular organization and heterogeneity. METHODS: We introduce a computational algorithm, termed Tumor-Immune Partitioning and Clustering (TIPC), that jointly measures immune cell partitioning between tumor epithelial and stromal areas and immune cell clustering versus dispersion. As proof-of-principle, we applied TIPC to a prospective cohort incident tumor biobank containing 931 colorectal carcinoma cases. TIPC identified tumor subtypes with unique spatial patterns between tumor cells and T lymphocytes linked to certain molecular pathologic and prognostic features. T lymphocyte identification and phenotyping were achieved using multiplexed (multispectral) immunofluorescence. In a separate hepatocellular carcinoma cohort, we replaced the stromal component with specific immune cell types-CXCR3+CD68+ or CD8+-to profile their spatial relationships with CXCL9+CD68+ cells. RESULTS: Six unsupervised TIPC subtypes based on T lymphocyte distribution patterns were identified, comprising two cold and four hot subtypes. Three of the four hot subtypes were associated with significantly longer colorectal cancer (CRC)-specific survival compared to a reference cold subtype. Our analysis showed that variations in T-cell densities among the TIPC subtypes did not strictly correlate with prognostic benefits, underscoring the prognostic significance of immune cell spatial patterns. Additionally, TIPC revealed two spatially distinct and cell density-specific subtypes among microsatellite instability-high colorectal cancers, indicating its potential to upgrade tumor subtyping. TIPC was also applied to additional immune cell types, eosinophils and neutrophils, identified using morphology and supervised machine learning; here two tumor subtypes with similarly low densities, namely 'cold, tumor-rich' and 'cold, stroma-rich', exhibited differential prognostic associations. Lastly, we validated our methods and results using The Cancer Genome Atlas colon and rectal adenocarcinoma data (n = 570). Moreover, applying TIPC to hepatocellular carcinoma cases (n = 27) highlighted critical cell interactions like CXCL9-CXCR3 and CXCL9-CD8. CONCLUSIONS: Unsupervised discoveries of microgeometric tissue organizational patterns and novel tumor subtypes using the TIPC algorithm can deepen our understanding of the tumor immune microenvironment and likely inform precision cancer immunotherapy.

Humans↗