SigMine and OPathDb: A Literature-Mining Pipeline and Opportunistic Pathogen Database

SigMine and OPathDb: A Literature-Mining Pipeline and Opportunistic Pathogen Database | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article SigMine and OPathDb: A Literature-Mining Pipeline and Opportunistic Pathogen Database Urvija Rani, Akshath Nair, Sonika Bhatnagar This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-9315983/v1 This work is licensed under a CC BY 4.0 License Status: Under Revision Version 1 posted 17 You are reading this latest preprint version Abstract Recognition of cross-domain association between biological entities in the vast biomedical literature is a challenging task. SigMine, an automated pipeline, was constructed to systematically mine biomedical literature to identify significantly associated biological entities. SigMine performs biomedical entity recognition from PMC articles using machine-learning and deep-learning–based entity recognition through Europe PMC Annotation API. Advanced entity recognition using Python scripting, NCBI E-Utilities, and an n-gram algorithm was performed followed by extensive data cleaning and mapping against standard databases. Statistical evaluation identified significant associations between entities. The entire workflow was automated through a modular framework developed in Python v3.13 with a Tkinter-based Graphical User Interface. SigMine enhances usability while retaining the flexibility to use new dictionaries for annotation. SigMine was used to construct a human Opportunistic Pathogens Database (OPathDb), housing 5,626 novel opportunistic pathogens significantly associated with 1,440 diseases and 7,121 genes mined from 25,000 PMC articles. Additional annotation of 598 significantly associated metabolites and 30 affected tissues is available for 3,204 and 227 pathogens respectively. OpathDb has a user-friendly query interface searchable by organism, disease, tissue, gene, protein and metabolite available at https://www.opathdb.cbsblab-nsut.in . Organism–entity associations can be visualized as weighted networks, with color-coded nodes and significance-scaled edges. Significant associations of opportunistic pathogens like Akkermansia mucinifila with colorectal cancer and Segatella copri with glucose intolerance can be identified through OpathDb. The SigMine framework demonstrates efficient recognition and prioritization of relationships in a vast and heterogenous corpora. biomedical literature mining opportunistic pathogens entity attribute significance associations database Figures Figure 1 Figure 2 Figure 3 Figure 4 1. Introduction A large number of biomedical research articles are published annually, making manual handling and curation difficult (Khatib Sulaiman Dalam No et al. 2024 ). Biomedical literature utilizes an array of synonymous terms to describe an entity or parts thereof. Much of the reported data from these research articles does not reflect in the associated databases. As a result, a substantial amount of implicit knowledge embedded in unstructured text remains inaccessible to computational analysis and systematic reuse. Therefore, advanced computational techniques using AI and natural language processing (NLP) have crucial applications in literature mining (Zhao et al. 2020 ). Prior studies have applied literature mining for specific biological associations. A literature mining tool integrates PubMed abstracts and genomic data to predict novel bone disease-associated genes using network mapping (Gajendran et al. 2007 ). A neural Bidirectional Long Short-Term Memory model was trained to extract detailed chronic obstructive pulmonary disease phenotypes by identifying in-text mentioned biomedical entities, such as proteins and treatments (Ju et al. 2019 ). A detailed analysis of the ketamine pathway and the role of gut microbiota in its antidepressant effects was performed using a hybrid approach that combined manual curation with automated named entity recognition and relation extraction to create pathway-specific knowledge graphs (Liu et al. 2023 ). Existing tools often have limited scope and are tailored for a limited number of biomedical parameters with limited scalability or flexibility. As an example, BioReader focuses on epitope-based information for communicable and non-communicable diseases (Simon et al. 2019 ). BioNerDs looks at usage trends of bioinformatics databases and tools (Duck et al. 2013 ). OmixLitMiner retrieves genomics and proteomics data related to cancer studies (Steffen et al. 2020 ). DTMiner identifies disease-gene associations (Xu et al. 2016 ), while MeInfoText identifies the methylation of genes in cancer (Fang et al. 2008 ). Textpresso extracts entities by categorizing sentences (Müller et al. 2004 ). iTextMine generates multi-entity knowledge graphs involving diseases, miRNAs, and drugs (Ren et al. 2018 ). Several literature mining frameworks have been developed to retrieve microbe-disease associations (Srivastava et al. 2019 ; Park et al. 2021 ; Wu et al. 2021 ). However, these frameworks do not extend to other biomedical entities like genes, proteins, tissues or metabolites or relationships. While standard APIs offered by the National Center for Biotechnology Information (NCBI) and Europe PubMed Central (PMC) facilitate large-scale literature annotation (Sayers 2022 ; Yang et al. 2023 ), there is no pipeline for integrative mining and annotation of biomedical literature for using these utilities and other annotated biological dictionaries. An automated user-friendly framework for large-scale, multi-entity literature mining that can be tailored to varying scientific questions and applications is required. Opportunistic pathogens plague hosts with impaired immune systems, such as immunodeficient individuals or people with underlying medical conditions (Fusco et al. 2018 ) while remaining harmless in healthy immunocompetent individuals. They are found in hospitals and healthcare centers (Hornef 2015 ) and can include bacteria, fungi, or viruses (José et al. 2020 ). The opportunistic pathogens can cause infection through mechanisms like colonization and adhesion of biological tissues or implanted medical devices using their surface proteins (Polvi et al. 2015 ), invasion of compromised physiological barrier during surgery, trauma, or mucositis (Polvi et al. 2015 ; Stringer et al. 2024 ), formation of biofilm on medical devices such as catheters and implants (Vuotto et al. 2014 ; Arciola et al. 2018 ), immune evasion (Youngs and Arnold 2021 ; Htun et al. 2024 ), toxin- and enzyme-mediated pathogenicity (Chebotar et al. 2025 ) and antibiotic resistance (Chen et al. 2025b ). Many opportunistic pathogens are reported in the literature. They are able to escape the immune system of the host by producing several virulence elements that assist them in avoiding identification and killing by immune cells (Hopkins et al. 2024 ). These factors enhance the ability of the pathogen to form a colony, destroy the host tissue, and subvert the immune response (Casadevall and Pirofski 2009 ). The National Healthcare Safety Network ( https://www.cdc.gov/nhsn/about-nhsn/index.html ) maintains a list of opportunistic pathogens in the human body. Developed by the Healthcare Infection Control Practices Advisory Committee, this list helps in tracking and reducing bloodstream infections in healthcare personnel. Existing databases such as gcPathogen (Guo et al. 2024 ), EuPathDB (Aurrecoechea et al. 2017 ), MPD (Zhang et al. 2018 ), Target-Pathogen (Sosa et al. 2017 ), and BacDive (Schober et al. 2025 ) are not specifically dedicated to opportunistic pathogens and mainly focus on genomic, phenotypic, or strain-level information. Additionally, many of the pathogen disease associations remain obscurely buried within the scientific literature, and can only be accessed by individually going through each and every publication painstakingly. Even then, the information may be completely missed even by a trained human eye. At present, there is no comprehensive database for opportunistic pathogens. In this study, we report a scalable and adaptive pipeline to mine literature-derived biological associations. SigMine utilizes extraction of biomedical entities through machine-learning and deep-learning–based entity recognition pipelines available via the Europe PMC Annotation Application Programming Interface (API). Other online resources are used for additional biomedical entity recognition. SigMine incorporates a systematic pre-processing approach followed by validation through cross-referencing with known biological databases. Further, entity attribute pairs occurring together in a large number of published articles are considered guilty by association. Application of chi-square testing with Bonferroni correction identifies significant associations among the annotated entity-attribute pairs. In the next stage, the entire workflow was fully automated through a modular software framework that also provides a Tkinter-based Graphical User Interface (GUI) to enable non-programmers to execute the pipeline seamlessly from data retrieval to statistical clustering. An opportunistic pathogen database, OPathDb, was developed to demonstrate the applicability of SigMine. OPathDb houses literature-mined opportunistic pathogens. OPathDb integrates 5,626 opportunistic pathogens linked to 1,440 diseases, 7,121 genes, 598 metabolites, and 30 tissues from ~ 25,000 PMC articles. OPathDb enables pathogen-centric analysis by linking novel opportunistic pathogens to associated diseases, tissues, genes/proteins, and metabolites, thereby facilitating the identification of virulence-related molecular factors and revealing tissue specificity or metabolite adaptations. The database is accessible publicly at https://www.opathdb.cbsblab-nsut.in . 2. Methodology Figure 1 illustrates the schematic representation of the methodology followed throughout the study. The following steps are involved from collection to curation and statistical testing using the pipeline: 2.1. Retrieval of full text articles A list of PMCIDs of the full-text biomedical literature is obtained from PMC database of NCBI using keyword-based queries. The Advanced Search option of PMC can be used to construct a judicious relevant query by combining keywords with Boolean operators. 2.2. Biomedical entity recognition This involves two sub-processes: Biomedical entity recognition using machine learning and deep learning-based text mining Europe PMC Annotation API : The retrieved PMCIDs are systematically annotated for biomedical entities, using the “annotationsByArticleIds” endpoint of the Europe PMC Annotation API ( https://europepmc.org/AnnotationsApi ) (Yang et al. 2023 ), which enables structured named entity annotations for full-text articles generated through automated text-mining pipelines incorporating machine-learning and deep-learning–based entity recognition methods. Annotations of organisms, diseases, genes, proteins, and chemicals were obtained using this resource. Biomedical entity recognition using other online resources : A custom sub-process integrating the NCBI E-Utilities, Python's BeautifulSoup library and N-gram algorithm for targeted web data extraction from additional libraries is used for biomedical entity recognition. An n-gram–based text-matching approach was used to support biomedical entity extraction by identifying contiguous sequences of n tokens in the text. This method performs detection of multi-word entity mentions that may not be captured by exact string matching. This lightweight, dictionary-driven process allows efficient identification of biomedical entities while maintaining scalability throughout large volumes of full-text literature. This method was used for tissue terms extraction from the full text articles by comparing them against the Foundational Model of Anatomy (FMA) dictionary (Rosse and Mejino 2003 ). However, the data retrieved using both the above entity recognition techniques contained numerous redundancies, was incomplete, incorrect, or non-standard. Therefore, further cleaning, standardization, and annotation were required. 2.3. Data pre-processing and cleaning Biomedical entities were pre-processed and cleaned using Python’s pandas library. This included dropping irrelevant columns like source, extId, annotations.prefix, annotations.postfix, annotations.sections, and annotations.id from the retrieved annotations. Null values and redundant entries were also removed from the dataset. 2.4. Curated database-driven data standardization Standardization of entity nomenclature and linking were performed by mapping extracted entities to authoritative biological databases and ontologies in: Organism Annotation : The annotations were standardized using taxonomic references from NCBI (Schoch et al. 2020 ) and UniprotKB (Bateman et al. 2025 ) databases to include only the desired classification and reject the incomplete taxonomic nomenclature. Retrieved organism annotations were mapped to corresponding NCBI TaxIDs to ensure consistent scientific nomenclature. Entries containing only genus-level classifications were removed to improve taxonomic resolution. Synonymous organism names (e.g., Escherichia coli and E. coli ) were standardized using TaxID mapping. Obsolete nomenclature (e.g., Propionibacterium acnes → Cutibacterium acnes ) and ambiguous/general terms that did not correspond to specific TaxIDs (e.g., bacteria, virus, fungi, plant, animal) were excluded. Disease Annotation : Retrieved disease annotations were mapped to the disease ontology provided by LinkedLife Data to standardize disease nomenclature (e.g., Colon cancer , CRC , and Colorectal carcinoma ) and remove disease terms without specific subtype classifications (e.g., cancer , diabetes ). In order to further filter the non-human diseases annotations, curated information was used from MalaCards (Espe 2018 ). MicroPhenoDB (Yao et al. 2021 ), Disbiome (Janssens et al. 2018 ), and Human Microbe-Disease Association Database (HMDAD) (Ma et al. 2017 ) was used to remove all known associated disease. Tissue Annotation : Tissue-related terms binned into organs and organ systems. This classification was performed by known tissue ontology-based dictionaries derived from the FMA and the BRENDA Tissue Ontology (BTO) (Gremse et al. 2011 ). Gene and Protein Annotation : Standardization of gene and protein entities was performed using UniprotKB (Bateman et al. 2025 ) database. Gene and protein entries were converted to standard gene (e.g., TP53, p53, and Tumor protein p53) and protein (e.g., IL6, IL-6, Interleukin-6) nomenclature using standard gene and protein names from UniprotKb. Gene family or common descriptors (e.g., MAP) that did not map to specific genes were removed. Similarly, protein family or gene symbol descriptors were removed from protein nomenclature (e.g., heat shock protein or HSP70) Metabolite Annotation : Metabolite entities standardized using Microbial Metabolome Database (MiMeDB) (Wishart et al. 2023 ). Retrieved metabolite annotations were mapped to standardize metabolite nomenclature (e.g., lactic acid and lactate ) using MiMeDB. Generalized metabolite descriptors that could not be mapped to specific metabolites (e.g., acid , base , alkaline ) were removed to retain only well-defined metabolite entities. Finally, all curated and pre-processed annotation datasets are consolidated using pandas DataFrame into an analysis-ready dataset. 2.5. Performing statistical analysis Entity associations were identified using co-occurrence analysis across full-text articles. A chi-square test was conducted on all annotated entity-attribute pairs in Python. Bonferroni adjustment was implemented to avoid Type I errors (false positives). In this step, organism was considered as an entity and associated disease. tissue, gene-protein, and metabolites as associated attributes. The entity–attribute associations with adjusted p-value < 0.05 were considered significant. Biomedical entity-centric clusters were generated based on varying levels of statistical significance. The significance level was classified into three groups: highly significant (adjusted p-value < 0.001), moderately significant (0.001 ≤ adjusted p-value < 0.01), and significant (adjusted p-value ≥ 0.01). Thus, generating clusters of different types such as having only highly, moderately, and significantly-related entities, and combinations of highly and moderately significant entities. 2.6. Merging of significant biomedical entity-attribute pairs Significant associations were merged using common entity identifiers (e.g., organisms) using pandas merge function. This results in an integrated multi-entity-attribute dataset. 2.7. Software framework implementation A custom software framework implementing the proposed pipeline was developed in Python (v3.13, 64-bit) with a Tkinter-based GUI to enable end-to-end automation of biomedical literature mining. The software is designed in a modular architecture, comprising distinct components for biomedical entity–attribute recognition, cleaning, standardization, statistical evaluation, and clustering. This modular design ensures reproducibility and ease of use. The framework of SigMine uses several open-source Python libraries for text mining, data processing, and statistical analysis, summarized in Supplementary Table S1 . 2.8. Database Architecture The OPathDB database was developed using a lightweight web-based architecture to aid data storage and retrieval of pathogen-associated biological information. The front-end of the database was developed using HTML, CSS, and JavaScript. The back-end of the database was developed in PHP to manage data retrieval and query processing. The standardized dataset was stored in a CSV file to ensure simplicity, portability, and ease of updates. PHP scripts dynamically parse this CSV file in response to user requests and return structured outputs in tabular form. 3. Results and Discussion SigMine is a pipeline implemented for large-scale biomedical entity-attribute mining from full-text biomedical literature, SigMine provides a systematic framework encompassing data retrieval, biomedical entity-attribute recognition, curated database-driven data cleaning, standardization, statistical analysis, and the generation of entity-centric clusters. 3.1. Functional workflow of the pipeline The tools used in the pipeline at every stage along with their purpose are summarized in Table 1 . The results of the different steps followed in the methodology section is described below: Table 1 Databases, Dictionaries and Tools Used in SigMine Pipeline for construction of OpathDb Step Databases/Dictionaries/Tools used Usage 1. Retrieval of full text articles NCBI PubMed Central (PMC) Retrieval of full-text articles using keyword queries 2. Machine-learning and deep-learning–driven Basic Entity extraction Europe PMC Annotation API Extraction of organism, disease, gene/protein, and metabolite entities 3. Python Pre-processing and standardization: Preprocessing by mapping with: a. Organism data cleaning NCBITaxID • Removal of redundant and null entities. • Removal of genus-level organism annotations. b. Disease nomenclature Disease Ontology using LinkedLife Data • Uniformity in disease nomenclature. c. Gene and protein nomenclature Uniprot database • Standardization of gene and protein names d. Metabolite nomenclature MiMeDB • Removal of generalized metabolite terms • Uniform metabolite nomenclature 4. Curated database-assisted standardization: Standardization by mapping with: a. Organism nomenclature UniProtKB • Standardization of organism annotations D • Discard unrelated organisms. b. Disease data cleaning MalaCards • To retain human-associated diseases c. Finding novel associations MicroPhenoDB, Disbiome, and HMDAD. • Removal of known microbe–disease associations Dictionary-driven data annotation a. Obtaining Tissue annotations • Retrieval from FMA ontology database using BeautifulSoup and NCBI E-Utilities • Mapping using NLP n-gram algorithm Identification and annotation of tissue-specific terms a. Cleaning Tissue annotations Mapping with FMA, BTO for categorization of tissues into organs and organ systems. • Removal of duplicate and null entries. • Binning affected tissues 5. Statistical evaluation and clustering Python (pandas, scipy) for chi-square testing • Evaluation of entity–attribute associations. • Entity-attribute clustering generation based Retrieval of full text articles: The keyword list was used to obtain the PMCIDs of the full text articles. 3.1.2. Biomedical entity recognition: This step was achieved in two stages: Biomedical entity recognition using machine-learning and deep learning–based text mining: Organisms, diseases, genes, proteins, and metabolites from each PMCID were obtained using Europe PMC Annotation API. Biomedical entity recognition using other online resources: For each of the publications, tissue terms were obtained as described in Methodology. 3.1.3. Data pre-processing and cleaning: Refined annotations included retained key columns corresponding to Pathogen, Disease, Tissue, Gene-Proteins, and Metabolites. This dataset was free from null and redundant entries. 3.1.4. Curated database-driven standardization: The retrieved annotations were further processed using Python-based pre-processing steps. This included: Organism Annotation: By mapping against standard databases, organism names were obtained in a standardized, well-defined genus species notation with synonymous terms unified, obsolete nomenclature updated, and incomplete genus-level/ambiguous taxonomic entries removed. Disease Annotation: A standardized disease nomenclature ensured that synonymous disease terms were unified while non-specific disease terms were removed. After annotation, the dataset included only human-associated diseases with novel microbe–disease associations. Tissue Annotation: Tissue annotations were obtained in a standardized anatomical terminology with entries categorized according to their organs and organ systems. Gene and Protein Annotation: Standardized gene and protein names were obtained. The synonymous terms were unified and ambiguous descriptors were discarded. Metabolite Annotation: Standardized Metabolite annotations were obtained while generalized entries not corresponding to specific metabolites were removed. A single dataframe is obtained after merging the standardized annotations from a single PMCID, thus eliminating the chance co-occurrence of named entities. This results in a single structured format of an entity with all its related attributes. To illustrate this, the separate annotations and their combined dataframe is shown in Supplementary Figure S1 . 3.1.5. Statistical evaluation and clustering: Significant organism–attribute associations were obtained on the basis of frequency of co-occurrence and classified into three categories; namely- highly significant, moderately significant, and significant. A single entity, e.g. organism is then selected as the center to which associated attributes are connected in order to highlight biologically relevant patterns. 3.1.6. Merging of significant biomedical entity-attribute pairs A unified, non-redundant multi-entity–attribute dataset comprising significant entity–attribute pairs based on the common entity selected is obtained. 3.2. SigMine: Software Adaptation The workflow of SigMine was obtained through a sequence of automated steps, as illustrated in Fig. 2 and Supplementary video. The implementation provided a GUI for the different modules. The Project manager module (Fig. 2 a and 2 b) allows the user to create a new project or open existing projects. This ensures the formation or opening of a single system folder containing all the files related to the project. Upon launching the project, the left-hand side panel shows the entire Project Pipeline consisting of Welcome, PMCID Manager, Filtering, Data Cleaning, Analysis and Results. The right-hand panel shows the main working interface, organized into six tabs: Welcome Tab: Confirmation of User ID/ Folder name (Fig. 2 c). PMCID Manager: Represents steps 2.1 and 2.2 of Methodology corresponding to input of PMCIDs and biomedical entity–attribute recognition. PMCIDs can uploaded using the input box and then be saved, cleared, or submitted for annotation using the Save , Clear and Submit buttons given below the input box as shown in Fig. 2 d. The annotated output is saved in the Project folder as multiple csv formatted files, one corresponding to each type of annotation obtained e.g. organism, disease, tissues, etc. Filtering: Corresponds to steps 2.3 or data pre-processing and cleaning stage of the Methodology. It is performed in the Filtering tab (Fig. 2 e). Multiple annotation files from the project folder are loaded automatically using the Check Annotations button. An Annotation information window shows the total number of annotations obtained from file. It is processed using the Filter Annotations button to remove null and redundant entries and the progress can be seen in the Filtering options window. A dialog box pos up and reports the process completion. The output is saved as multiple cleaned files in the project folder. Data Cleaning: Corresponds to Step 2.4 described in the Methodology section or curated database-based standardization and is illustrated Fig. 2 f. Multiple filtered annotation files are loaded using the Check Filtered Annotations button and standardized against curated databases using the Clean Data button. The entries uploaded can be seen in the Filtered annotation window while the progress can be monitored through the cleaning options window. The cleaned datasets are merged into a single dataframe using the Merge button (Fig. 2 f). Popup dialogue boxes report successful completion at every stage. The output is saved as a single csv file. Analysis: Fig. 2 g depicts the step 2.5 or Statistical evaluation stage of the methodology. The entity–attribute pairs are analyzed using a configurable p-value threshold via the Run Analysis button. The output is saved as multiple files containing each entity attribute pair and the corresponding p-value (lower than the cutoff specified). Results: Represents the step 2.6 or merging of significant entity-attribute pairs stage of the methodology (Fig. 2 h) using the Categorize & Merge Results button. The final output is then saved as a CSV file in the project directory. 3.3. Comparison of SigMine with other methods Most existing literature-mining tools focus on specific tasks such as disease–gene association (e.g., DTMiner), phenotype annotation (e.g., Ju et al., 2019 for COPD phenotype), or document classification (e.g., BioReader), and are often limited to abstracts or curated datasets (e.g., MeInfoText). The majority of these methods do not support multi-entity integration and lack statistical validation. A comparison of SigMine with other available methods is shown in Table 2 . As evident from the Table, SigMine supports multiple biological entities, enables integration across different data types, and includes statistical analysis for identifying significant associations. It also provides a modular workflow that allows flexible data input and processing, making it more comprehensive in comparison with existing approaches. Table 2 Comparison Of SigMine With Existing Literature-Mining Approaches Tool / Study Primary focus Input scope Supported entities Multi-entity integration Statistical validation Automation level Gajendran et al., 2007 Disease-gene discovery (bone biology) PubMed abstracts + genomics Genes, disease No No Semi-automated Ju et al., 2019 (COPD phenotypes) Phenotype annotation Curated corpus Disease phenotypes Limited No Semi-automated Liu et al., 2023 Pathway-centric knowledge graph Literature + pathway DBs Drugs, microbes, neurotransmitters Partial No Semi-automated BioReader Document classification Abstracts Text categories No No Automated BioNerDS Tool/database usage trends Abstracts Tools, databases No No Automated OmixLitMiner Omics lead prioritization Literature + omics data Genes, proteins No No Semi-automated DTMiner Disease-gene associations Abstracts Disease, gene No No Automated MeInfoText Gene methylation in cancer Abstracts Genes, methylation No No Automated Textpresso Ontology-based entity extraction Full text Entity mentions No No Semi-automated iTextMine Knowledge graph construction Full text Disease, miRNA, drugs Partial No Automated Park et al., 2021 Microbe–disease associations Abstracts Microbe, disease No No Automated (DL-based) Wu et al., 2021 Microbe–disease associations Abstracts Microbe, disease No No Automated (DL-based) EviMass Human microbial associations Abstracts Microbe, disease No Fisher test Semi-automated SigMine (this study) Generalized literature mining Full-text PMC Organisms, diseases, tissues, genes/proteins, metabolites Yes Yes (χ² + Bonferroni) Fully automated 3.4. Use case: Literature mining of human opportunistic pathogens The pipeline was applied to mine human opportunistic pathogens from the PMC database, and uncover their biologically significant associations. The pipeline was implemented using the list of PMCIDs obtained with the keywords “human” AND “opportunistic pathogen”. As a result, a total of 25,425 research papers on human opportunistic pathogens were obtained from the NCBI PMC database. Using the PMCIDs obtained, annotations for organisms, diseases, genes, proteins, metabolites, and tissues were retrieved. Data cleaning and standardization were performed as described in the methodology to ensure high-quality data integration. Significant associations between opportunistic pathogens and their associated attributes were obtained through statistical analysis and are summarized in Table 3 . Associated entity–attribute pairs according to different levels of statistical significance were determined and are shown in Table 4 . Clusters containing only highly significant associations represent the strongest relationships between opportunistic pathogens and their associated attributes, with gene/protein associations showing the highest representation. Table 3 Counts of the Statistically Significant Biomedical Entity-Attribute Associations Number of Opportunistic Pathogens Number and Types of Associated Attributes Corrected P-value range 5626 1440 Disease 3.87 × 10⁻³⁰⁰ – 0.04996 227 30 Tissue 6.81 × 10⁻⁶⁶ – 0.04995 6788 7121 Gene-Protein 3.00 × 10⁻³⁰⁵ – 0.04995 3204 598 Metabolite 2.94 × 10⁻²⁹⁸ – 0.04987 Table 4 Organism-Centric Clusters of Opportunistic Pathogens at Different Significance Levels Cluster Type Opportunistic Pathogens Associated Diseases Associated Tissues Associated Genes/Proteins Associated Metabolites All highly significant entities 5183 1342 25 6391 551 4 highly significant + 1 moderately significant 5183 1342 25 6391 231 3 highly significant + 2 moderately significant 5183 1342 25 2335 231 2 highly significant + 3 moderately significant 5183 1342 21 2335 231 All moderately significant entities 1111 640 9 1457 105 Only significant entities 1006 584 14 876 104 Significance thresholds: highly significant (adjusted p-value < 0.001), moderately significant (0.001 ≤ adjusted p-value < 0.01), significant (adjusted p-value ≥ 0.01). OPathDb provides a web interface for exploring significantly associated opportunistic pathogen–based relationships identified through the biomedical literature mining pipeline, as illustrated in Fig. 3 . The database contains information on opportunistic pathogens and their associated diseases, tissues, genes/proteins, and metabolites. The home page includes an overview of the database and a search form for querying the data (Fig. 3 a and 3 b). The search module provides access to different entity categories, including organism, disease, tissue, gene/protein, and metabolite. This enables retrieval of opportunistic pathogen–associated information. Keyword-based retrieval of entity-specific data is shown in the search interface, such as an organism name in the Organism category (Fig. 3 c). The retrieved results are displayed in a structured tabular format (Fig. 3 d). Organism-centric multipartite network visualizations representing significant pathogen–attribute associations were obtained in OPathDb (Fig. 4 ). This network depicts the pathogen as the central node, with its associated diseases, genes/proteins, metabolites, and tissues as connected nodes. Different node colors indicate different biological entity types, whereas edge colors and weights represent levels of statistical significance (3 = high, 2 = moderate, 1 = significant). Additional information regarding database functionality and usage is available in the integrated FAQ/Tutorial section (Supplementary Table S2). 3.5. Comparison OpathDb data with other sources In order to validate OPathDb against established pathogen data repositories, a pathogen-entry-based comparison was performed against the NHSN organism list ( https://www.cdc.gov/nhsn/about-nhsn/index.html ) and the gcPathogen database (Guo et al. 2024 ). Using direct pathogen mapping, OpathDb matched 924 out of 2,278 NHSN-listed organisms (40.56% coverage) and 428 out of 858 gcPathogen organisms (49.88% coverage). Importantly, pathogens that do not overlap with these reference resources are literature-derived entries uniquely identified by the SigMine pipeline. This highlights the ability of SigMine to capture novel entity associations that are not explicitly catalogued in existing surveillance lists or pathogen databases. 3.6. Case study-based validation of OpathDb The correctness of the data in OPathDb was checked by examining the details of prominent categories of opportunistic pathogens as found in literature. Table 5 summarizes representative opportunistic pathogens and their highly significant mined associations in OPathDb. Gastrointestinal taxa including Akkermansia muciniphila (Zheng et al. 2023 ; Gubernatorova et al. 2023 ; Siwo et al. 2024 ; Zhang et al. 2025 ), Bacteroides fragilis (Sánchez et al. 2012 ), and Segatella copri (Abdelsalam et al. 2023 ; Xiao et al. 2024b , a ; Zhou et al. 2025 ) were associated with colorectal cancer, celiac disease, and glucose intolerance, respectively, consistent with existing literature. Oral pathogens such as Fusobacterium nucleatum (Liu et al. 2014 ; Dahlstrand Rudin et al. 2021 ; Chen et al. 2022a , b ; Wang et al. 2023 ) and Treponema denticola (Nakagawa et al. 2002 ; Listyarifah et al. 2018 ; Zang et al. 2022 ) were linked to periodontal and oral cancer contexts, accompanied by known adhesins, stress-response proteins, and inflammatory metabolites as substantiated by previous studies. Cutaneous pathogens including Cutibacterium acne (Won et al. 2022 ; Sauer et al. 2024 ; Chen et al. 2025a ; Burma and Ramien 2025 ), Enterococcus faecium (Sharifi et al. 2013 ; Goel et al. 2016 ; Arshadi et al. 2018 ; Codelia-Anjum et al. 2023 ; Wu et al. 2024 ), Klebsiella pneumoniae (Wiskur et al. 2008 ; Sridhar et al. 2014 ; Serban et al. 2021 ; Ren et al. 2023 ), and Pseudomonas aeruginosa (Koh et al. 2005 ; Woods et al. 2011 ; Nesher et al. 2015 ; Qin et al. 2022 ) were associated with alopecia areata, urinary tract infection, endophthalmitis, and neutropenia, respectively, as stated in previous studies. Organisms uniquely present in OpathDb but not in the NHSN list and gcPathogen database included A. muciniphila, S. copri and T. denticola . Collectively, these examples indicate that OPathDb accurately consolidates literature-supported opportunistic pathogen–attribute relationships throughout diverse tissues while capturing relevant molecular and metabolic signatures. Table 5 Case study–based validation of highly significant opportunistic pathogen–attributes associations captured in OpathDb Opportunistic Pathogen Associated Disease Associated Tissue Associated Gene & Protein Associated Metabolite References Akkermansia muciniphila Colorectal Cancer Gastrointestinal Tract TLR2, NLRP3, LPS, TLR4, MUC2, FMT Lactic acid [50–53] Bacteroides fragilis Celiac Disease Gastrointestinal Tract TLR2, TLR4, NOD2, MUC2 Inulin [54] Cutibacterium acnes Alopecia Areata Skin RECA, COL11A2, COL9A2, COL9A3 Azelaic acid [67–70] Enterococcus faecium Urinary Tract Infection Urinary System Acm, Ebp, Ef1 Colistin, Tigecycline [71–75] Fusobacterium nucleatum Aggressive Periontidis Mouth CXCL9, CXCL10, CXCL16,CDH17, GAPDH, JNK Butyric acid [59–63] Klebsiella pneumoniae Endophthalmitis Urinary System Cps Aerobactin [76–79] Pseudomonas aeruginosa Neutropenia Faeces ExsA, LasA, ToxA, PilA, OprF, AlgD, FliC, PopB, Rhl, Ly6G, CFTR 2-heptyl-3-hydroxy-4-quinolone [80–83] Segatella copri Glucose Intolerance Faeces BSH, DCA, LBP Phenylacetylglutamine, Sedoheptulose 7-phosphate, Trimethylamine [55–58] Treponema denticola Tongue squamous cell carcinoma Mouth KGP, LPS, FimA Butyric acid [64–66] 4. Conclusion This study presents SigMine, a scalable and adaptive literature-mining pipeline designed to systematically process large volumes of unstructured biomedical text in the context of structured biological knowledge to reliably elucidate significant associations among biological entities. The pipeline integrates entity recognition using machine-learning and deep learning–based text mining along with other online-based resources, followed by statistical evaluation to identify significant associations. The framework is readily adaptable for generating domain-specific knowledgebases to support clinical research, surveillance, and evidence-driven decision-making. Thus, this study makes a meaningful contribution toward addressing significant literature mining challenges in computational biology by developing an automated scalable pipeline that: (a) efficiently handles large-scale entity annotation and association discovery, (b) integrates multiple bioinformatics tools, databases and standardized ontologies for robust curation, (c) statistically validates results for biological significance, d) implemented in a database, and e) provides modularity with intermittent input/output, enabling the user to upload data from any source to be cleaned, standardized, annotated, and subjected to merging, statistical tests, and multi-entity clustering. . While remaining flexible for extension via dictionary and ontology updates, SigMine can be easily adapted for large-scale literature mining to address a wide range of associations such as host-pathogen, disease-disease, disease-gene, disease-metabolite, drug-pathway, drug-disease, drug-metabolite, drug–target interactions, and pathway-level relationships. SigMine was applied as a use case to mine human opportunistic pathogens and their associated diseases, tissues, genes, and metabolites. This resulted in the development of OPathDb, a curated database of pathogen-centric associations. The pipeline enables the generation of pathogen–attribute clustered by statistical significance levels that can be viewed in tabular form or graphically in OPathDb. The current annotation framework focuses on organisms, diseases, tissues, genes/proteins, and metabolites but provides a flexible basis for further expansion. It can be extended to include additional biomedical entities and relationships. This enhances its application for large-scale knowledge discovery and supporting varied uses in biomedical and clinical research. Abbreviations API, Application Programming Interface; BTO, BRENDA Tissue Ontology; CSV, Comma-Separated Values; CSS, Cascading Style Sheet; FMA, Foundational Model of Anatomy; GUI, Graphical User Interface; HMDAD, Human Microbe-Disease Association Database; HMDB, Human Metabolome Database; HTML, Hypertext Markup Language; MiMeDB, Microbial Metabolome Database; NCBI, National Center of Biotechnology Information; NLP, Natural Language Processing; PHP, Hypertext Preprocessor; PMC, PubMed Central Declarations Funding: UR acknowledges research support from Netaji Subhas University of Technology, New Delhi, through a grant of University Research Fellowship. SB acknowledges support from DBT project BT/PR40197/BTIS/137/68/2023. Competing Interest: The authors have no relevant financial or non-financial interests to disclose. Author Contribution: Urvija Rani: Methodology, Formal Analysis and Writing; Akshath Nair: Software Implementation; Sonika Bhatnagar: Conceptualization,Supervision, Project management and Review ORCiD Information Urvija Rani: https://orcid.org/0000-0001-5520-8691 Sonika Bhatnagar: https://orcid.org/0000-0002-8818-4240 Acknowledgements: We gratefully acknowledge the funding and use of High Performance Computing Facility from NSUT, New Delhi. Data Availability The data generated is available in a the OPathDb database (https://www.opathdb.cbsblab-nsut.in). References Abdelsalam NA, Hegazy SM, Aziz RK (2023) The curious case of Prevotella copri. Gut Microbes 15:. https://doi.org/10.1080/19490976.2023.2249152 Arciola CR, Campoccia D, Montanaro L (2018) Implant infections: adhesion, biofilm formation and immune evasion. Nature Reviews Microbiology 2018 16:7 16:397–409. https://doi.org/10.1038/s41579-018-0019-y Arshadi M, Mahmoudi M, Motahar MS, et al (2018) Virulence Determinants and Antimicrobial Resistance Patterns of Vancomycin-resistant Enterococcus faecium Isolated from Different Sources in Southwest Iran. Iran J Public Health 47:264 Aurrecoechea C, Barreto A, Basenko EY, et al (2017) EuPathDB: the eukaryotic pathogen genomics database resource. Nucleic Acids Res 45:D581–D591. https://doi.org/10.1093/NAR/GKW1105 Bateman A, Martin MJ, Orchard S, et al (2025) UniProt: the Universal Protein Knowledgebase in 2025. Nucleic Acids Res 53:D609–D617. https://doi.org/10.1093/NAR/GKAE1010 Burma NE, Ramien ML (2025) Cutaneous and Gut Dysbiosis in Alopecia Areata: A Review. JID Innov 5:100363. https://doi.org/10.1016/j.xjidi.2025.100363 Casadevall A, Pirofski LA (2009) Virulence factors and their mechanisms of action: the view from a damage–response framework. J Water Health 7:S2–S18. https://doi.org/10.2166/WH.2009.036 Chebotar I, Azizov I, Edelstein M, Kozlov R (2025) Detection of virulence factors in opportunistic bacteria: advances, challenges, and practical implementation. Front Microbiol 16:1638925. https://doi.org/10.3389/FMICB.2025.1638925/FULL Chen Q, Liu C, Tao J, et al (2025a) Insights into microbial dysbiosis and Cutibacterium acnes CAMP factor interactions in acne vulgaris. Microb Genom 11:001449. https://doi.org/10.1099/mgen.0.001449 Chen S, Liao L, Wang M (2025b) Editorial: Opportunistic pathogens: pathogenesis and multi-drug resistance mechanisms. Front Microbiol 16:1597769. https://doi.org/10.3389/FMICB.2025.1597769/BIBTEX Chen Y, Huang Z, Tang Z, et al (2022a) More Than Just a Periodontal Pathogen –the Research Progress on Fusobacterium nucleatum. Front Cell Infect Microbiol 12:815318. https://doi.org/10.3389/fcimb.2022.815318 Chen Y, Shi T, Li Y, et al (2022b) Fusobacterium nucleatum: The Opportunistic Pathogen of Periodontal and Peri-Implant Diseases. Front Microbiol 13:860149. https://doi.org/10.3389/FMICB.2022.860149/XML Codelia-Anjum A, Lerner LB, Elterman D, et al (2023) Enterococcal Urinary Tract Infections: A Review of the Pathogenicity, Epidemiology, and Treatment. Antibiotics 12:778. https://doi.org/10.3390/antibiotics12040778 Dahlstrand Rudin A, Khamzeh A, Venkatakrishnan V, et al (2021) Short chain fatty acids released by Fusobacterium nucleatum are neutrophil chemoattractants acting via free fatty acid receptor 2 (FFAR2). Cell Microbiol 23:e13348. https://doi.org/10.1111/cmi.13348 Duck G, Nenadic G, Brass A, et al (2013) BioNerDS: Exploring bioinformatics’ database and software use through literature mining. BMC Bioinformatics 14:1–13. https://doi.org/10.1186/1471-2105-14-194/FIGURES/7 Espe S (2018) Malacards: The Human Disease Database. J Med Libr Assoc 106:140. https://doi.org/10.5195/JMLA.2018.253 Fang YC, Huang HC, Juan HF (2008) MeInfoText: associated gene methylation and cancer information from text mining. BMC Bioinformatics 9:22. https://doi.org/10.1186/1471-2105-9-22 Fusco V, Abriouel H, Benomar N, et al (2018) Opportunistic Food-Borne Pathogens. Food Safety and Preservation 269–306. https://doi.org/10.1016/B978-0-12-814956-0.00010-X Gajendran VK, Lin JR, Fyhrie DP (2007) An application of bioinformatics and text mining to the discovery of novel genes related to bone biology. Bone 40:1378–1388. https://doi.org/10.1016/J.BONE.2006.12.067 Goel V, Kumar D, Kumar R, et al (2016) Community Acquired Enterococcal Urinary Tract Infections and Antibiotic Resistance Profile in North India. J Lab Physicians 8:50. https://doi.org/10.4103/0974-2727.176237 Gremse M, Chang A, Schomburg I, et al (2011) The BRENDA Tissue Ontology (BTO): the first all-integrating ontology of all organisms for enzyme sources. Nucleic Acids Res 39:D507–D513. https://doi.org/10.1093/NAR/GKQ968 Gubernatorova EO, Gorshkova EA, Bondareva MA, et al (2023) Akkermansia muciniphila - friend or foe in colorectal cancer? Front Immunol 14:1303795. https://doi.org/10.3389/FIMMU.2023.1303795/BIBTEX Guo C, Chen Q, Fan G, et al (2024) gcPathogen: a comprehensive genomic resource of human pathogens for public health. Nucleic Acids Res 52:D714–D723. https://doi.org/10.1093/NAR/GKAD875 Hopkins HA, Lopezguerra C, Lau MJ, Raymann K (2024) Making a Pathogen? Evaluating the Impact of Protist Predation on the Evolution of Virulence in Serratia marcescens. Genome Biol Evol 16:. https://doi.org/10.1093/GBE/EVAE149 Hornef M (2015) Pathogens, Commensal Symbionts, and Pathobionts: Discovery and Functional Effects on the Host. ILAR J 56:159–162. https://doi.org/10.1093/ILAR/ILV007 Htun ZM, Gul MH, Sadikot RT (2024) Bacterial Infections in Patients Living with HIV. Results Probl Cell Differ 73:537. https://doi.org/10.1007/978-3-031-62036-2_21 Janssens Y, Nielandt J, Bronselaer A, et al (2018) Disbiome database: Linking the microbiome to disease. BMC Microbiol 18:1–6. https://doi.org/10.1186/S12866-018-1197-5/TABLES/1 José RJ, Periselneris JN, Brown JS (2020) Opportunistic bacterial, viral and fungal infections of the lung. Medicine 48:366–372. https://doi.org/10.1016/J.MPMED.2020.03.006 Ju M, Short AD, Thompson P, et al (2019) Annotating and detecting phenotypic information for chronic obstructive pulmonary disease. JAMIA Open 2:261. https://doi.org/10.1093/JAMIAOPEN/OOZ009 Khatib Sulaiman Dalam No J, Atam Mustafa A, Mohsin Abdulazeez A (2024) A Review of Text Classification Based on ML & Data Mining Algorithms. The Indonesian Journal of Computer Science 13:. https://doi.org/10.33022/IJCS.V13I3.4027 Koh AY, Priebe GP, Pier GB (2005) Virulence of Pseudomonas aeruginosa in a Murine Model of Gastrointestinal Colonization and Dissemination in Neutropenia. Infect Immun 73:2262. https://doi.org/10.1128/IAI.73.4.2262-2272.2005 Listyarifah D, Nieminen MT, Mäkinen LK, et al (2018) Treponema denticola chymotrypsin-like proteinase is present in early-stage mobile tongue squamous cell carcinoma and related to the clinicopathological features. Journal of Oral Pathology and Medicine 47:764–772. https://doi.org/10.1111/jop.12729 Liu P, Liu Y, Wang J, et al (2014) Detection of Fusobacterium Nucleatum and fadA Adhesin Gene in Patients with Orthodontic Gingivitis and Non-Orthodontic Periodontal Inflammation. PLoS One 9:e85280. https://doi.org/10.1371/journal.pone.0085280 Liu T, Feenstra KA, Huang Z, Heringa J (2023) Mining literature and pathway data to explore the relations of ketamine with neurotransmitters and gut microbiota using a knowledge-graph. Bioinformatics 40:btad771. https://doi.org/10.1093/BIOINFORMATICS/BTAD771 Ma W, Zhang L, Zeng P, et al (2017) An analysis of human microbe–disease associations. Brief Bioinform 18:85–97. https://doi.org/10.1093/BIB/BBW005 Müller HM, Kenny EE, Sternberg PW (2004) Textpresso: An Ontology-Based Information Retrieval and Extraction System for Biological Literature. PLoS Biol 2:e309. https://doi.org/10.1371/JOURNAL.PBIO.0020309 Nakagawa I, Amano A, Kuboniwa M, et al (2002) Functional Differences among FimA Variants of Porphyromonas gingivalis and Their Effects on Adhesion to and Invasion of Human Epithelial Cells. Infect Immun 70:277. https://doi.org/10.1128/IAI.70.1.277-285.2002 Nesher L, Rolston KVI, Shah DP, et al (2015) Fecal colonization and infection with Pseudomonas aeruginosa in recipients of allogeneic hematopoietic stem cell transplantation. Transplant Infectious Disease 17:33–38. https://doi.org/10.1111/tid.12323 Park Y, Lee J, Moon H, et al (2021) Discovering microbe-disease associations from the literature using a hierarchical long short-term memory network and an ensemble parser model. Scientific Reports 2021 11:1 11:1–12. https://doi.org/10.1038/s41598-021-83966-8 Polvi EJ, Li X, O’Meara TR, et al (2015) Opportunistic yeast pathogens: reservoirs, virulence mechanisms, and therapeutic strategies. Cell Mol Life Sci 72:2261. https://doi.org/10.1007/S00018-015-1860-Z Qin S, Xiao W, Zhou C, et al (2022) Pseudomonas aeruginosa: pathogenesis, virulence factors, antibiotic resistance, interaction with host, technology advances and emerging therapeutics. Signal Transduction and Targeted Therapy 2022 7:1 7:1–27. https://doi.org/10.1038/s41392-022-01056-1 Ren C, Li Z, Meng F, et al (2023) Endogenous endophthalmitis caused by urinary tract infection: A case report. Medicine (United States) 102:E36139. https://doi.org/10.1097/MD.0000000000036139 Ren J, Li G, Ross K, et al (2018) iTextMine: integrated text-mining system for large-scale knowledge extraction from the literature. Database (Oxford) 2018:bay128. https://doi.org/10.1093/DATABASE/BAY128 Rosse C, Mejino JLV (2003) A reference ontology for biomedical informatics: the Foundational Model of Anatomy. J Biomed Inform 36:478–500. https://doi.org/10.1016/J.JBI.2003.11.007 Sánchez E, Laparra JM, Sanz Y (2012) Discerning the Role of Bacteroides fragilis in Celiac Disease Pathogenesis. Appl Environ Microbiol 78:6507. https://doi.org/10.1128/AEM.00563-12 Sauer N, Oślizło M, Brzostek M, et al (2024) The multiple uses of azelaic acid in dermatology: mechanism of action, preparations, and potential therapeutic applications. Advances in Dermatology and Allergology/Postȩpy Dermatologii i Alergologii 40:716. https://doi.org/10.5114/ada.2023.133955 Sayers E (2022) A General Introduction to the E-utilities Schober I, Koblitz J, Carbasse JS, et al (2025) BacDive in 2025: the core database for prokaryotic strain data. Nucleic Acids Res 53:D748–D756. https://doi.org/10.1093/NAR/GKAE959 Schoch CL, Ciufo S, Domrachev M, et al (2020) NCBI Taxonomy: a comprehensive update on curation, resources and tools. Database (Oxford) 2020:baaa062. https://doi.org/10.1093/DATABASE/BAAA062 Serban D, Popa Cherecheanu A, Dascalu AM, et al (2021) Hypervirulent Klebsiella pneumoniae Endogenous Endophthalmitis—A Global Emerging Disease. 11:676. https://doi.org/10.3390/life11070676 Sharifi Y, Hasani A, Ghotaslou R, et al (2013) Virulence and Antimicrobial Resistance in Enterococci Isolated from Urinary Tract Infections. Adv Pharm Bull 3:197. https://doi.org/10.5681/apb.2013.032 Simon C, Davidsen K, Hansen C, et al (2019) BioReader: A text mining tool for performing classification of biomedical literature. BMC Bioinformatics 19:165–170. https://doi.org/10.1186/S12859-019-2607-X/TABLES/1 Siwo GH, Singal AG, Waljee AK (2024) Pan-cancer molecular signatures connecting aspartate transaminase (AST) to cancer prognosis, metabolic and immune signatures. bioRxiv 2024.03.01.582939. https://doi.org/10.1101/2024.03.01.582939 Sosa EJ, Burguener G, Lanzarotti E, et al (2017) Target-Pathogen: a structural bioinformatic approach to prioritize drug targets in pathogens. Nucleic Acids Res 46:D413. https://doi.org/10.1093/NAR/GKX1015 Sridhar J, Flynn HW, Kuriyan AE, et al (2014) Endophthalmitis caused by Klebsiella species. Retina 34:1875. https://doi.org/10.1097/IAE.0000000000000162 Srivastava D, Baksi KD, Kuntal BK, Mande SS (2019) “EviMass”: A Literature Evidence-Based Miner for Human Microbial Associations. Front Genet 10:466029. https://doi.org/10.3389/FGENE.2019.00849/BIBTEX Steffen P, Wu J, Hariharan S, et al (2020) OmixLitMiner: A Bioinformatics Tool for Prioritizing Biological Leads from ‘Omics Data Using Literature Retrieval and Data Mining. International Journal of Molecular Sciences 2020, Vol 21, Page 1374 21:1374. https://doi.org/10.3390/IJMS21041374 Stringer AM, Hargreaves BM, Mendes RA, et al (2024) Updated perspectives on the contribution of the microbiome to the pathogenesis of mucositis using the MASCC/ISOO framework. Supportive Care in Cancer 32:558. https://doi.org/10.1007/S00520-024-08752-4 Vuotto C, Longo F, Balice MP, et al (2014) Antibiotic Resistance Related to Biofilm Formation in Klebsiella pneumoniae. Pathogens 2014, Vol 3, Pages 743-758 3:743–758. https://doi.org/10.3390/PATHOGENS3030743 Wang Y, Wang L, Sun T, et al (2023) Study of the inflammatory activating process in the early stage of Fusobacterium nucleatum infected PDLSCs. International Journal of Oral Science 2023 15:1 15:8-. https://doi.org/10.1038/s41368-022-00213-0 Wishart DS, Oler E, Peters H, et al (2023) MiMeDB: the Human Microbial Metabolome Database. Nucleic Acids Res 51:D611–D620. https://doi.org/10.1093/NAR/GKAC868 Wiskur BJ, Hunt JJ, Callegan MC (2008) Hypermucoviscosity as a Virulence Factor in Experimental Klebsiella pneumoniae Endophthalmitis. Invest Ophthalmol Vis Sci 49:4931. https://doi.org/10.1167/iovs.08-2276 Won EJ, Jang HH, Park H, Kim SJ (2022) A Potential Predictive Role of the Scalp Microbiome Profiling in Patients with Alopecia Areata: Staphylococcus caprae, Corynebacterium, and Cutibacterium Species. Microorganisms 10:864. https://doi.org/10.3390/microorganisms10050864 Woods DE, Lam JS, Paranchych W, et al (2011) Correlation of Pseudomonas aeruginosa virulence factors from clinical and environmental isolates with pathogenicity in the neutropenic mouse. https://doi.org/101139/m97-077 43:541–551. https://doi.org/10.1139/m97-077 Wu C, Xiao X, Yang C, et al (2021) Mining microbe–disease interactions from literature via a transfer learning model. BMC Bioinformatics 22:1–15. https://doi.org/10.1186/S12859-021-04346-7/TABLES/3 Wu W, Xiao S, Han L, Wu Q (2024) Antimicrobial resistance, virulence gene profiles, and molecular epidemiology of enterococcal isolates from patients with urinary tract infections in Shanghai, China. Microbiol Spectr 13:e01217-24. https://doi.org/10.1128/spectrum.01217-24 Xiao X, Singh A, Giometto A, Brito IL (2024a) Segatella clades adopt distinct roles within a single individual’s gut. NPJ Biofilms Microbiomes 10:114. https://doi.org/10.1038/s41522-024-00590-w Xiao X, Singh A, Giometto A, Brito IL (2024b) Segatella copri strains adopt distinct roles within a single individual’s gut. bioRxiv 2024.05.20.595015. https://doi.org/10.1101/2024.05.20.595015 Xu D, Zhang M, Xie Y, et al (2016) DTMiner: identification of potential disease targets through biomedical literature mining. Bioinformatics 32:3619–3626. https://doi.org/10.1093/BIOINFORMATICS/BTW503 Yang X, Saha S, Venkatesan A, et al (2023) Europe PMC annotated full-text corpus for gene/proteins, diseases and organisms. Scientific Data 2023 10:1 10:1–13. https://doi.org/10.1038/s41597-023-02617-x Yao G, Zhang W, Yang M, et al (2021) MicroPhenoDB Associates Metagenomic Data with Pathogenic Microbes, Microbial Core Genes, and Human Disease Phenotypes. Genomics Proteomics Bioinformatics 18:760. https://doi.org/10.1016/J.GPB.2020.11.001 Youngs J, Arnold A (2021) Infections in the immunocompromised host: secondary immunodeficiency disorders. Medicine 49:611–617. https://doi.org/10.1016/J.MPMED.2021.07.003 Zang W, Liu J, Geng F, et al (2022) Butyrate promotes oral squamous cell carcinoma cells migration, invasion and epithelial-mesenchymal transition. PeerJ 10:e12991. https://doi.org/10.7717/peerj.12991 Zhang K, Zhu YW, Tang AQ, et al (2025) Role of 3-mercaptopyruvate sulfurtransferase in cancer: Molecular mechanisms and therapeutic perspectives. Transl Oncol 52:102272. https://doi.org/10.1016/j.tranon.2025.102272 Zhang T, Miao J, Han N, et al (2018) MPD: a pathogen genome and metagenome database. Database 2018:. https://doi.org/10.1093/DATABASE/BAY055 Zhao S, Su C, Lu Z, Wang F (2020) Recent advances in biomedical literature mining. Brief Bioinform 22:bbaa057. https://doi.org/10.1093/BIB/BBAA057 Zheng M, Han R, Yuan Y, et al (2023) The role of Akkermansia muciniphila in inflammatory bowel disease: Current knowledge and perspectives. Front Immunol 13:1089600. https://doi.org/10.3389/FIMMU.2022.1089600/BIBTEX Zhou C, Hibberd MC, Lee EM, et al (2025) Glycoside hydrolase–mediated glucomannan catabolism in Segatella copri, a target of microbiota-directed foods for malnourished children. Proc Natl Acad Sci U S A 122:e2521522122. https://doi.org/10.1073/pnas.2521522122 Additional Declarations No competing interests reported. Supplementary Files SupplementaryFile.docx Cite Share Download PDF Status: Under Revision Version 1 posted Editorial decision: Revision requested 19 Apr, 2026 Reviews received at journal 19 Apr, 2026 Reviews received at journal 16 Apr, 2026 Reviews received at journal 15 Apr, 2026 Reviews received at journal 14 Apr, 2026 Reviewers agreed at journal 12 Apr, 2026 Reviews received at journal 11 Apr, 2026 Reviewers agreed at journal 09 Apr, 2026 Reviewers agreed at journal 08 Apr, 2026 Reviewers agreed at journal 07 Apr, 2026 Reviewers agreed at journal 07 Apr, 2026 Reviewers agreed at journal 07 Apr, 2026 Reviewers agreed at journal 07 Apr, 2026 Reviewers invited by journal 07 Apr, 2026 Editor assigned by journal 07 Apr, 2026 Submission checks completed at journal 06 Apr, 2026 First submitted to journal 03 Apr, 2026 You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-9315983","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":621614193,"identity":"a7908dd1-4af9-4f3d-9f76-f778ee30e6fe","order_by":0,"name":"Urvija Rani","email":"","orcid":"","institution":"Netaji Subhas University of Technology","correspondingAuthor":false,"prefix":"","firstName":"Urvija","middleName":"","lastName":"Rani","suffix":""},{"id":621614194,"identity":"e5bc7b84-d3c2-437f-93cb-9d6ce7ad2dbf","order_by":1,"name":"Akshath Nair","email":"","orcid":"","institution":"Uppsala University","correspondingAuthor":false,"prefix":"","firstName":"Akshath","middleName":"","lastName":"Nair","suffix":""},{"id":621614195,"identity":"ce575d8e-4d41-4954-9da4-40ea1a8f38c9","order_by":2,"name":"Sonika Bhatnagar","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA8ElEQVRIiWNgGAWjYDACCR4Yi/kAXPAAFoWoWiAq2BKAhAFJWngM4FrwAvnZvQc/f9xhZ9c/+8y3hz/b/jDwtx9gPFyAR4vBnXPJEgfPJCfPOJe73Zi3zYBB4kwCw+EZ+LRI5BhIHGxjTmY4w7tNmhGoheEGA8NhHjxa5GfkGP842FafLH+G55nkT6AWeUJaGG7kmAFtOWxncIaHTQLkMANCWgzunDGzONt2PMHwDJuZNM85Yx7DM4kN+B02u8f4RmVbtb3cGeZnkj/K5OTkjh8+/Bmvw6AgsQHKACpmbMCjEAHsiVI1CkbBKBgFIxMAAHK5TKkpJONSAAAAAElFTkSuQmCC","orcid":"","institution":"Netaji Subhas University of Technology","correspondingAuthor":true,"prefix":"","firstName":"Sonika","middleName":"","lastName":"Bhatnagar","suffix":""}],"badges":[],"createdAt":"2026-04-03 20:08:33","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-9315983/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-9315983/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":106815411,"identity":"b066ab89-f987-4091-b36b-2d8b7ed11f85","added_by":"auto","created_at":"2026-04-13 17:10:48","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":622596,"visible":true,"origin":"","legend":"\u003cp\u003eSchematic representation of the proposed literature mining pipeline. (a) Retrieval of biomedical literature and extraction of biological entities. (b) Data cleaning, standardization, annotation, statistical evaluation, and clustering of extracted entities. (c) SigMine implementation and construction of the OPATHDB.\u003c/p\u003e","description":"","filename":"Figure1.png","url":"https://assets-eu.researchsquare.com/files/rs-9315983/v1/f0730c5979a8fc23b826c9e9.png"},{"id":106815399,"identity":"5d21fbe8-1824-4b85-9eea-7ec07ac54664","added_by":"auto","created_at":"2026-04-13 17:10:46","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":400823,"visible":true,"origin":"","legend":"\u003cp\u003eWorkflow of SigMine. (a) SigMine launch page. (b) User login to create a new project or open an existing project. (c) Welcome window of SigMine (d) Uploading PMCIDs via the \u003cem\u003eUpload PMCIDs\u003c/em\u003e button, followed by annotation retrieval using the \u003cem\u003eFetch Annotation\u003c/em\u003e button. (e) Loading and filtering annotations on the \u003cem\u003eFiltering\u003c/em\u003epage using the \u003cem\u003eCheck Annotation\u003c/em\u003e button. (f) Loading and cleaning annotations on the \u003cem\u003eData Cleaning\u003c/em\u003e page using the \u003cem\u003eCheck Annotation\u003c/em\u003ebutton. Merging cleaned annotations using the \u003cem\u003eClean and Merge\u003c/em\u003e button. (g) Performing statistical analysis on the \u003cem\u003eAnalysis\u003c/em\u003e page via the \u003cem\u003eRun Analysis\u003c/em\u003e button. (h) Consolidating and categorizing significant entity pairs on the \u003cem\u003eResult\u003c/em\u003e page using the \u003cem\u003eCategorize and Merge Results\u003c/em\u003ebutton.\u003c/p\u003e","description":"","filename":"Figure2.png","url":"https://assets-eu.researchsquare.com/files/rs-9315983/v1/e711a5f3b950ca21a9dab12f.png"},{"id":106815428,"identity":"0f44058d-0b30-4d4c-a7e9-6b0a47675c08","added_by":"auto","created_at":"2026-04-13 17:10:51","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":587221,"visible":true,"origin":"","legend":"\u003cp\u003eUser interface and search workflow of OPathDb. (a) Home page of OPathDb. (b) Searching by different category types. (c) Keyword-based search for the Organism category. (d) Display of search result.\u003c/p\u003e","description":"","filename":"Figure3.png","url":"https://assets-eu.researchsquare.com/files/rs-9315983/v1/d3735658d5fc6f4964e23c39.png"},{"id":106815424,"identity":"3eccd518-d4a6-4950-a63b-45326cac4118","added_by":"auto","created_at":"2026-04-13 17:10:49","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":506029,"visible":true,"origin":"","legend":"\u003cp\u003eOrganism-centric multipartite network showing significant associations between opportunistic pathogens (red nodes) and their attributes: diseases (blue nodes), tissues (green nodes), genes/proteins (yellow nodes), and metabolites (purple nodes). Edge colors indicate significance levels: blue (highly significant, weight = 3), purple (moderately significant, weight = 2), and black (significant, weight = 1).\u003c/p\u003e","description":"","filename":"Figure4.png","url":"https://assets-eu.researchsquare.com/files/rs-9315983/v1/c047ed87c5a26abf91c22081.png"},{"id":106960910,"identity":"c4ab6528-ef10-4370-8100-4055c2e63c9c","added_by":"auto","created_at":"2026-04-15 09:23:36","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":3230808,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-9315983/v1/4758a14d-31bb-4532-938a-06d8dca3d091.pdf"},{"id":106815414,"identity":"56f5b5bf-4de2-493f-98a5-a9c4cb7456ff","added_by":"auto","created_at":"2026-04-13 17:10:49","extension":"docx","order_by":5,"title":"","display":"","copyAsset":false,"role":"supplement","size":968265,"visible":true,"origin":"","legend":"","description":"","filename":"SupplementaryFile.docx","url":"https://assets-eu.researchsquare.com/files/rs-9315983/v1/bb58265a1f8d74f5afc99da1.docx"}],"financialInterests":"No competing interests reported.","formattedTitle":"SigMine and OPathDb: A Literature-Mining Pipeline and Opportunistic Pathogen Database","fulltext":[{"header":"1. Introduction","content":"\u003cp\u003eA large number of biomedical research articles are published annually, making manual handling and curation difficult (Khatib Sulaiman Dalam No et al. \u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e2024\u003c/span\u003e). Biomedical literature utilizes an array of synonymous terms to describe an entity or parts thereof. Much of the reported data from these research articles does not reflect in the associated databases. As a result, a substantial amount of implicit knowledge embedded in unstructured text remains inaccessible to computational analysis and systematic reuse. Therefore, advanced computational techniques using AI and natural language processing (NLP) have crucial applications in literature mining (Zhao et al. \u003cspan citationid=\"CR76\" class=\"CitationRef\"\u003e2020\u003c/span\u003e).\u003c/p\u003e \u003cp\u003ePrior studies have applied literature mining for specific biological associations. A literature mining tool integrates PubMed abstracts and genomic data to predict novel bone disease-associated genes using network mapping (Gajendran et al. \u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e2007\u003c/span\u003e). A neural Bidirectional Long Short-Term Memory model was trained to extract detailed chronic obstructive pulmonary disease phenotypes by identifying in-text mentioned biomedical entities, such as proteins and treatments (Ju et al. \u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e2019\u003c/span\u003e). A detailed analysis of the ketamine pathway and the role of gut microbiota in its antidepressant effects was performed using a hybrid approach that combined manual curation with automated named entity recognition and relation extraction to create pathway-specific knowledge graphs (Liu et al. \u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e2023\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eExisting tools often have limited scope and are tailored for a limited number of biomedical parameters with limited scalability or flexibility. As an example, BioReader focuses on epitope-based information for communicable and non-communicable diseases (Simon et al. \u003cspan citationid=\"CR52\" class=\"CitationRef\"\u003e2019\u003c/span\u003e). BioNerDs looks at usage trends of bioinformatics databases and tools (Duck et al. \u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e2013\u003c/span\u003e). OmixLitMiner retrieves genomics and proteomics data related to cancer studies (Steffen et al. \u003cspan citationid=\"CR57\" class=\"CitationRef\"\u003e2020\u003c/span\u003e). DTMiner identifies disease-gene associations (Xu et al. \u003cspan citationid=\"CR69\" class=\"CitationRef\"\u003e2016\u003c/span\u003e), while MeInfoText identifies the methylation of genes in cancer (Fang et al. \u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e2008\u003c/span\u003e). Textpresso extracts entities by categorizing sentences (M\u0026uuml;ller et al. \u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e2004\u003c/span\u003e). iTextMine generates multi-entity knowledge graphs involving diseases, miRNAs, and drugs (Ren et al. \u003cspan citationid=\"CR43\" class=\"CitationRef\"\u003e2018\u003c/span\u003e). Several literature mining frameworks have been developed to retrieve microbe-disease associations (Srivastava et al. \u003cspan citationid=\"CR56\" class=\"CitationRef\"\u003e2019\u003c/span\u003e; Park et al. \u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e2021\u003c/span\u003e; Wu et al. \u003cspan citationid=\"CR65\" class=\"CitationRef\"\u003e2021\u003c/span\u003e). However, these frameworks do not extend to other biomedical entities like genes, proteins, tissues or metabolites or relationships. While standard APIs offered by the National Center for Biotechnology Information (NCBI) and Europe PubMed Central (PMC) facilitate large-scale literature annotation (Sayers \u003cspan citationid=\"CR47\" class=\"CitationRef\"\u003e2022\u003c/span\u003e; Yang et al. \u003cspan citationid=\"CR70\" class=\"CitationRef\"\u003e2023\u003c/span\u003e), there is no pipeline for integrative mining and annotation of biomedical literature for using these utilities and other annotated biological dictionaries. An automated user-friendly framework for large-scale, multi-entity literature mining that can be tailored to varying scientific questions and applications is required.\u003c/p\u003e \u003cp\u003eOpportunistic pathogens plague hosts with impaired immune systems, such as immunodeficient individuals or people with underlying medical conditions (Fusco et al. \u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e2018\u003c/span\u003e) while remaining harmless in healthy immunocompetent individuals. They are found in hospitals and healthcare centers (Hornef \u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e2015\u003c/span\u003e) and can include bacteria, fungi, or viruses (Jos\u0026eacute; et al. \u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e2020\u003c/span\u003e). The opportunistic pathogens can cause infection through mechanisms like colonization and adhesion of biological tissues or implanted medical devices using their surface proteins (Polvi et al. \u003cspan citationid=\"CR40\" class=\"CitationRef\"\u003e2015\u003c/span\u003e), invasion of compromised physiological barrier during surgery, trauma, or mucositis (Polvi et al. \u003cspan citationid=\"CR40\" class=\"CitationRef\"\u003e2015\u003c/span\u003e; Stringer et al. \u003cspan citationid=\"CR58\" class=\"CitationRef\"\u003e2024\u003c/span\u003e), formation of biofilm on medical devices such as catheters and implants (Vuotto et al. \u003cspan citationid=\"CR59\" class=\"CitationRef\"\u003e2014\u003c/span\u003e; Arciola et al. \u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2018\u003c/span\u003e), immune evasion (Youngs and Arnold \u003cspan citationid=\"CR72\" class=\"CitationRef\"\u003e2021\u003c/span\u003e; Htun et al. \u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e2024\u003c/span\u003e), toxin- and enzyme-mediated pathogenicity (Chebotar et al. \u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e2025\u003c/span\u003e) and antibiotic resistance (Chen et al. \u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e2025b\u003c/span\u003e). Many opportunistic pathogens are reported in the literature. They are able to escape the immune system of the host by producing several virulence elements that assist them in avoiding identification and killing by immune cells (Hopkins et al. \u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e2024\u003c/span\u003e). These factors enhance the ability of the pathogen to form a colony, destroy the host tissue, and subvert the immune response (Casadevall and Pirofski \u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e2009\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eThe National Healthcare Safety Network (\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://www.cdc.gov/nhsn/about-nhsn/index.html\u003c/span\u003e\u003cspan address=\"https://www.cdc.gov/nhsn/about-nhsn/index.html\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e) maintains a list of opportunistic pathogens in the human body. Developed by the Healthcare Infection Control Practices Advisory Committee, this list helps in tracking and reducing bloodstream infections in healthcare personnel. Existing databases such as gcPathogen (Guo et al. \u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e2024\u003c/span\u003e), EuPathDB (Aurrecoechea et al. \u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e2017\u003c/span\u003e), MPD (Zhang et al. \u003cspan citationid=\"CR75\" class=\"CitationRef\"\u003e2018\u003c/span\u003e), Target-Pathogen (Sosa et al. \u003cspan citationid=\"CR54\" class=\"CitationRef\"\u003e2017\u003c/span\u003e), and BacDive (Schober et al. \u003cspan citationid=\"CR48\" class=\"CitationRef\"\u003e2025\u003c/span\u003e) are not specifically dedicated to opportunistic pathogens and mainly focus on genomic, phenotypic, or strain-level information. Additionally, many of the pathogen disease associations remain obscurely buried within the scientific literature, and can only be accessed by individually going through each and every publication painstakingly. Even then, the information may be completely missed even by a trained human eye. At present, there is no comprehensive database for opportunistic pathogens.\u003c/p\u003e \u003cp\u003eIn this study, we report a scalable and adaptive pipeline to mine literature-derived biological associations. SigMine utilizes extraction of biomedical entities through machine-learning and deep-learning\u0026ndash;based entity recognition pipelines available via the Europe PMC Annotation Application Programming Interface (API). Other online resources are used for additional biomedical entity recognition. SigMine incorporates a systematic pre-processing approach followed by validation through cross-referencing with known biological databases. Further, entity attribute pairs occurring together in a large number of published articles are considered guilty by association. Application of chi-square testing with Bonferroni correction identifies significant associations among the annotated entity-attribute pairs.\u003c/p\u003e \u003cp\u003eIn the next stage, the entire workflow was fully automated through a modular software framework that also provides a Tkinter-based Graphical User Interface (GUI) to enable non-programmers to execute the pipeline seamlessly from data retrieval to statistical clustering. An opportunistic pathogen database, OPathDb, was developed to demonstrate the applicability of SigMine. OPathDb houses literature-mined opportunistic pathogens. OPathDb integrates 5,626 opportunistic pathogens linked to 1,440 diseases, 7,121 genes, 598 metabolites, and 30 tissues from ~\u0026thinsp;25,000 PMC articles. OPathDb enables pathogen-centric analysis by linking novel opportunistic pathogens to associated diseases, tissues, genes/proteins, and metabolites, thereby facilitating the identification of virulence-related molecular factors and revealing tissue specificity or metabolite adaptations. The database is accessible publicly at \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://www.opathdb.cbsblab-nsut.in\u003c/span\u003e\u003cspan address=\"https://www.opathdb.cbsblab-nsut.in\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/p\u003e"},{"header":"2. Methodology","content":"\u003cp\u003eFigure\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e illustrates the schematic representation of the methodology followed throughout the study. The following steps are involved from collection to curation and statistical testing using the pipeline:\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cdiv id=\"Sec3\" class=\"Section2\"\u003e \u003ch2\u003e2.1. Retrieval of full text articles\u003c/h2\u003e \u003cp\u003eA list of PMCIDs of the full-text biomedical literature is obtained from PMC database of NCBI using keyword-based queries. The Advanced Search option of PMC can be used to construct a judicious relevant query by combining keywords with Boolean operators.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec4\" class=\"Section2\"\u003e \u003ch2\u003e2.2. Biomedical entity recognition\u003c/h2\u003e \u003cp\u003eThis involves two sub-processes:\u003c/p\u003e \u003cp\u003e \u003cul\u003e \u003cli\u003e \u003cp\u003e \u003cem\u003eBiomedical entity recognition using machine learning and deep learning-based text mining Europe PMC Annotation API\u003c/em\u003e:\u003c/p\u003e \u003c/li\u003e \u003c/ul\u003e \u003c/p\u003e \u003cp\u003eThe retrieved PMCIDs are systematically annotated for biomedical entities, using the \u0026ldquo;annotationsByArticleIds\u0026rdquo; endpoint of the Europe PMC Annotation API (\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://europepmc.org/AnnotationsApi\u003c/span\u003e\u003cspan address=\"https://europepmc.org/AnnotationsApi\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e) (Yang et al. \u003cspan citationid=\"CR70\" class=\"CitationRef\"\u003e2023\u003c/span\u003e), which enables structured named entity annotations for full-text articles generated through automated text-mining pipelines incorporating machine-learning and deep-learning\u0026ndash;based entity recognition methods. Annotations of organisms, diseases, genes, proteins, and chemicals were obtained using this resource.\u003c/p\u003e \u003cp\u003e \u003cul\u003e \u003cli\u003e \u003cp\u003e \u003cem\u003eBiomedical entity recognition using other online resources\u003c/em\u003e:\u003c/p\u003e \u003c/li\u003e \u003c/ul\u003e \u003c/p\u003e \u003cp\u003eA custom sub-process integrating the NCBI E-Utilities, Python's \u003cem\u003eBeautifulSoup\u003c/em\u003e library and N-gram algorithm for targeted web data extraction from additional libraries is used for biomedical entity recognition. An n-gram\u0026ndash;based text-matching approach was used to support biomedical entity extraction by identifying contiguous sequences of \u003cem\u003en\u003c/em\u003e tokens in the text. This method performs detection of multi-word entity mentions that may not be captured by exact string matching. This lightweight, dictionary-driven process allows efficient identification of biomedical entities while maintaining scalability throughout large volumes of full-text literature. This method was used for tissue terms extraction from the full text articles by comparing them against the Foundational Model of Anatomy (FMA) dictionary (Rosse and Mejino \u003cspan citationid=\"CR44\" class=\"CitationRef\"\u003e2003\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eHowever, the data retrieved using both the above entity recognition techniques contained numerous redundancies, was incomplete, incorrect, or non-standard. Therefore, further cleaning, standardization, and annotation were required.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec5\" class=\"Section2\"\u003e \u003ch2\u003e2.3. Data pre-processing and cleaning\u003c/h2\u003e \u003cp\u003eBiomedical entities were pre-processed and cleaned using Python\u0026rsquo;s \u003cem\u003epandas\u003c/em\u003e library. This included dropping irrelevant columns like source, extId, annotations.prefix, annotations.postfix, annotations.sections, and annotations.id from the retrieved annotations. Null values and redundant entries were also removed from the dataset.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec6\" class=\"Section2\"\u003e \u003ch2\u003e2.4. Curated database-driven data standardization\u003c/h2\u003e \u003cp\u003eStandardization of entity nomenclature and linking were performed by mapping extracted entities to authoritative biological databases and ontologies in:\u003c/p\u003e \u003cp\u003e \u003cul\u003e \u003cli\u003e \u003cp\u003e \u003cem\u003eOrganism Annotation\u003c/em\u003e: The annotations were standardized using taxonomic references from NCBI (Schoch et al. \u003cspan citationid=\"CR49\" class=\"CitationRef\"\u003e2020\u003c/span\u003e) and UniprotKB (Bateman et al. \u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e2025\u003c/span\u003e) databases to include only the desired classification and reject the incomplete taxonomic nomenclature. Retrieved organism annotations were mapped to corresponding NCBI TaxIDs to ensure consistent scientific nomenclature. Entries containing only genus-level classifications were removed to improve taxonomic resolution. Synonymous organism names (e.g., \u003cem\u003eEscherichia coli\u003c/em\u003e and \u003cem\u003eE. coli\u003c/em\u003e) were standardized using TaxID mapping. Obsolete nomenclature (e.g., \u003cem\u003ePropionibacterium acnes\u003c/em\u003e \u0026rarr; \u003cem\u003eCutibacterium acnes\u003c/em\u003e) and ambiguous/general terms that did not correspond to specific TaxIDs (e.g., bacteria, virus, fungi, plant, animal) were excluded.\u003c/p\u003e \u003c/li\u003e \u003cli\u003e \u003cp\u003e \u003cem\u003eDisease Annotation\u003c/em\u003e: Retrieved disease annotations were mapped to the disease ontology provided by LinkedLife Data to standardize disease nomenclature (e.g., \u003cem\u003eColon cancer\u003c/em\u003e, \u003cem\u003eCRC\u003c/em\u003e, and \u003cem\u003eColorectal carcinoma\u003c/em\u003e) and remove disease terms without specific subtype classifications (e.g., \u003cem\u003ecancer\u003c/em\u003e, \u003cem\u003ediabetes\u003c/em\u003e). In order to further filter the non-human diseases annotations, curated information was used from MalaCards (Espe \u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e2018\u003c/span\u003e). MicroPhenoDB (Yao et al. \u003cspan citationid=\"CR71\" class=\"CitationRef\"\u003e2021\u003c/span\u003e), Disbiome (Janssens et al. \u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e2018\u003c/span\u003e), and Human Microbe-Disease Association Database (HMDAD) (Ma et al. \u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e2017\u003c/span\u003e) was used to remove all known associated disease.\u003c/p\u003e \u003c/li\u003e \u003cli\u003e \u003cp\u003e \u003cem\u003eTissue Annotation\u003c/em\u003e: Tissue-related terms binned into organs and organ systems. This classification was performed by known tissue ontology-based dictionaries derived from the FMA and the BRENDA Tissue Ontology (BTO) (Gremse et al. \u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e2011\u003c/span\u003e).\u003c/p\u003e \u003c/li\u003e \u003cli\u003e \u003cp\u003e \u003cem\u003eGene and Protein Annotation\u003c/em\u003e: Standardization of gene and protein entities was performed using UniprotKB (Bateman et al. \u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e2025\u003c/span\u003e) database. Gene and protein entries were converted to standard gene (e.g., TP53, p53, and Tumor protein p53) and protein (e.g., IL6, IL-6, Interleukin-6) nomenclature using standard gene and protein names from UniprotKb. Gene family or common descriptors (e.g., MAP) that did not map to specific genes were removed. Similarly, protein family or gene symbol descriptors were removed from protein nomenclature (e.g., heat shock protein or HSP70)\u003c/p\u003e \u003c/li\u003e \u003cli\u003e \u003cp\u003e \u003cem\u003eMetabolite Annotation\u003c/em\u003e: Metabolite entities standardized using Microbial Metabolome Database (MiMeDB) (Wishart et al. \u003cspan citationid=\"CR61\" class=\"CitationRef\"\u003e2023\u003c/span\u003e). Retrieved metabolite annotations were mapped to standardize metabolite nomenclature (e.g., \u003cem\u003elactic acid\u003c/em\u003e and \u003cem\u003elactate\u003c/em\u003e) using MiMeDB. Generalized metabolite descriptors that could not be mapped to specific metabolites (e.g., \u003cem\u003eacid\u003c/em\u003e, \u003cem\u003ebase\u003c/em\u003e, \u003cem\u003ealkaline\u003c/em\u003e) were removed to retain only well-defined metabolite entities.\u003c/p\u003e \u003c/li\u003e \u003cli\u003e \u003cp\u003eFinally, all curated and pre-processed annotation datasets are consolidated using \u003cem\u003epandas\u003c/em\u003e DataFrame into an analysis-ready dataset.\u003c/p\u003e \u003c/li\u003e \u003c/ul\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec7\" class=\"Section2\"\u003e \u003ch2\u003e2.5. Performing statistical analysis\u003c/h2\u003e \u003cp\u003eEntity associations were identified using co-occurrence analysis across full-text articles. A chi-square test was conducted on all annotated entity-attribute pairs in Python. Bonferroni adjustment was implemented to avoid Type I errors (false positives). In this step, organism was considered as an entity and associated disease. tissue, gene-protein, and metabolites as associated attributes. The entity\u0026ndash;attribute associations with adjusted p-value\u0026thinsp;\u0026lt;\u0026thinsp;0.05 were considered significant.\u003c/p\u003e \u003cp\u003eBiomedical entity-centric clusters were generated based on varying levels of statistical significance. The significance level was classified into three groups: highly significant (adjusted \u003cem\u003ep-value\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.001), moderately significant (0.001\u0026thinsp;\u0026le;\u0026thinsp;adjusted \u003cem\u003ep-value\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.01), and significant (adjusted \u003cem\u003ep-value\u003c/em\u003e\u0026thinsp;\u0026ge;\u0026thinsp;0.01). Thus, generating clusters of different types such as having only highly, moderately, and significantly-related entities, and combinations of highly and moderately significant entities.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec8\" class=\"Section2\"\u003e \u003ch2\u003e2.6. Merging of significant biomedical entity-attribute pairs\u003c/h2\u003e \u003cp\u003eSignificant associations were merged using common entity identifiers (e.g., organisms) using pandas merge function. This results in an integrated multi-entity-attribute dataset.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec9\" class=\"Section2\"\u003e \u003ch2\u003e2.7. Software framework implementation\u003c/h2\u003e \u003cp\u003eA custom software framework implementing the proposed pipeline was developed in Python (v3.13, 64-bit) with a Tkinter-based GUI to enable end-to-end automation of biomedical literature mining. The software is designed in a modular architecture, comprising distinct components for biomedical entity\u0026ndash;attribute recognition, cleaning, standardization, statistical evaluation, and clustering. This modular design ensures reproducibility and ease of use. The framework of SigMine uses several open-source Python libraries for text mining, data processing, and statistical analysis, summarized in Supplementary Table \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003e.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec10\" class=\"Section2\"\u003e \u003ch2\u003e2.8. Database Architecture\u003c/h2\u003e \u003cp\u003eThe OPathDB database was developed using a lightweight web-based architecture to aid data storage and retrieval of pathogen-associated biological information. The front-end of the database was developed using HTML, CSS, and JavaScript. The back-end of the database was developed in PHP to manage data retrieval and query processing. The standardized dataset was stored in a CSV file to ensure simplicity, portability, and ease of updates. PHP scripts dynamically parse this CSV file in response to user requests and return structured outputs in tabular form.\u003c/p\u003e \u003c/div\u003e"},{"header":"3. Results and Discussion","content":"\u003cp\u003eSigMine is a pipeline implemented for large-scale biomedical entity-attribute mining from full-text biomedical literature, SigMine provides a systematic framework encompassing data retrieval, biomedical entity-attribute recognition, curated database-driven data cleaning, standardization, statistical analysis, and the generation of entity-centric clusters.\u003c/p\u003e \u003cdiv id=\"Sec12\" class=\"Section2\"\u003e \u003ch2\u003e3.1. Functional workflow of the pipeline\u003c/h2\u003e \u003cp\u003eThe tools used in the pipeline at every stage along with their purpose are summarized in Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e. The results of the different steps followed in the methodology section is described below:\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab1\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 1\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eDatabases, Dictionaries and Tools Used in SigMine Pipeline for construction of OpathDb\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"3\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eStep\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eDatabases/Dictionaries/Tools used\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eUsage\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e1. Retrieval of full text articles\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eNCBI PubMed Central (PMC)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eRetrieval of full-text articles using keyword queries\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e2. Machine-learning and deep-learning\u0026ndash;driven Basic Entity extraction\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eEurope PMC Annotation API\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eExtraction of organism, disease, gene/protein, and metabolite entities\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colspan=\"3\" nameend=\"c3\" namest=\"c1\"\u003e \u003cp\u003e3. Python Pre-processing and standardization: Preprocessing by mapping with:\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ea. Organism data cleaning\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eNCBITaxID\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e\u0026bull; Removal of redundant and null entities.\u003c/p\u003e \u003cp\u003e\u0026bull; Removal of genus-level organism annotations.\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eb. Disease nomenclature\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eDisease Ontology using LinkedLife Data\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e\u0026bull; Uniformity in disease nomenclature.\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ec. Gene and protein nomenclature\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eUniprot database\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e\u0026bull; Standardization of gene and protein names\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ed. Metabolite nomenclature\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eMiMeDB\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e\u0026bull; Removal of generalized metabolite terms\u003c/p\u003e \u003cp\u003e\u0026bull; Uniform metabolite nomenclature\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colspan=\"3\" nameend=\"c3\" namest=\"c1\"\u003e \u003cp\u003e4. Curated database-assisted standardization: Standardization by mapping with:\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ea. Organism nomenclature\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eUniProtKB\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e\u0026bull; Standardization of organism annotations D\u003c/p\u003e \u003cp\u003e\u0026bull; Discard unrelated organisms.\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eb. Disease data cleaning\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eMalaCards\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e\u0026bull; To retain human-associated diseases\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ec. Finding novel associations\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eMicroPhenoDB, Disbiome, and HMDAD.\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e\u0026bull; Removal of known microbe\u0026ndash;disease associations\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colspan=\"3\" nameend=\"c3\" namest=\"c1\"\u003e \u003cp\u003eDictionary-driven data annotation\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ea. Obtaining Tissue annotations\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e\u0026bull; Retrieval from FMA ontology database using BeautifulSoup and NCBI E-Utilities\u003c/p\u003e \u003cp\u003e\u0026bull; Mapping using NLP n-gram algorithm\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eIdentification and annotation of tissue-specific terms\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ea. Cleaning Tissue annotations\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eMapping with FMA, BTO for categorization of tissues into organs and organ systems.\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e\u0026bull; Removal of duplicate and null entries.\u003c/p\u003e \u003cp\u003e\u0026bull; Binning affected tissues\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e5. Statistical evaluation and clustering\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003ePython (pandas, scipy) for chi-square testing\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e\u0026bull; Evaluation of entity\u0026ndash;attribute associations.\u003c/p\u003e \u003cp\u003e\u0026bull; Entity-attribute clustering generation based\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003e \u003col\u003e \u003cspan\u003e \u003cli\u003e \u003cp\u003eRetrieval of full text articles: The keyword list was used to obtain the PMCIDs of the full text articles.\u003c/p\u003e \u003c/li\u003e \u003c/span\u003e \u003c/ol\u003e \u003c/p\u003e \u003cdiv id=\"Sec13\" class=\"Section3\"\u003e \u003ch2\u003e3.1.2. Biomedical entity recognition: This step was achieved in two stages:\u003c/h2\u003e \u003cp\u003e \u003cul\u003e \u003cli\u003e \u003cp\u003eBiomedical entity recognition using machine-learning and deep learning\u0026ndash;based text mining: Organisms, diseases, genes, proteins, and metabolites from each PMCID were obtained using Europe PMC Annotation API.\u003c/p\u003e \u003c/li\u003e \u003cli\u003e \u003cp\u003eBiomedical entity recognition using other online resources: For each of the publications, tissue terms were obtained as described in Methodology.\u003c/p\u003e \u003c/li\u003e \u003c/ul\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec14\" class=\"Section3\"\u003e \u003ch2\u003e3.1.3. Data pre-processing and cleaning:\u003c/h2\u003e \u003cp\u003e \u003cdiv class=\"BlockQuote\"\u003e \u003cp\u003eRefined annotations included retained key columns corresponding to Pathogen, Disease, Tissue, Gene-Proteins, and Metabolites. This dataset was free from null and redundant entries.\u003c/p\u003e \u003c/div\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec15\" class=\"Section3\"\u003e \u003ch2\u003e3.1.4. Curated database-driven standardization: The retrieved annotations were further processed using Python-based pre-processing steps. This included:\u003c/h2\u003e \u003cp\u003e \u003cul\u003e \u003cli\u003e \u003cp\u003eOrganism Annotation: By mapping against standard databases, organism names were obtained in a standardized, well-defined genus species notation with synonymous terms unified, obsolete nomenclature updated, and incomplete genus-level/ambiguous taxonomic entries removed.\u003c/p\u003e \u003c/li\u003e \u003cli\u003e \u003cp\u003eDisease Annotation: A standardized disease nomenclature ensured that synonymous disease terms were unified while non-specific disease terms were removed. After annotation, the dataset included only human-associated diseases with novel microbe\u0026ndash;disease associations.\u003c/p\u003e \u003c/li\u003e \u003cli\u003e \u003cp\u003eTissue Annotation: Tissue annotations were obtained in a standardized anatomical terminology with entries categorized according to their organs and organ systems.\u003c/p\u003e \u003c/li\u003e \u003cli\u003e \u003cp\u003eGene and Protein Annotation: Standardized gene and protein names were obtained. The synonymous terms were unified and ambiguous descriptors were discarded.\u003c/p\u003e \u003c/li\u003e \u003cli\u003e \u003cp\u003eMetabolite Annotation: Standardized Metabolite annotations were obtained while generalized entries not corresponding to specific metabolites were removed.\u003c/p\u003e \u003c/li\u003e \u003cli\u003e \u003cp\u003eA single dataframe is obtained after merging the standardized annotations from a single PMCID, thus eliminating the chance co-occurrence of named entities. This results in a single structured format of an entity with all its related attributes. To illustrate this, the separate annotations and their combined dataframe is shown in Supplementary Figure \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003e.\u003c/p\u003e \u003c/li\u003e \u003c/ul\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec16\" class=\"Section3\"\u003e \u003ch2\u003e3.1.5. Statistical evaluation and clustering:\u003c/h2\u003e \u003cp\u003eSignificant organism\u0026ndash;attribute associations were obtained on the basis of frequency of co-occurrence and classified into three categories; namely- highly significant, moderately significant, and significant. A single entity, e.g. organism is then selected as the center to which associated attributes are connected in order to highlight biologically relevant patterns.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec17\" class=\"Section3\"\u003e \u003ch2\u003e3.1.6. Merging of significant biomedical entity-attribute pairs\u003c/h2\u003e \u003cp\u003e \u003cdiv class=\"BlockQuote\"\u003e \u003cp\u003eA unified, non-redundant multi-entity\u0026ndash;attribute dataset comprising significant entity\u0026ndash;attribute pairs based on the common entity selected is obtained.\u003c/p\u003e \u003c/div\u003e \u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv id=\"Sec18\" class=\"Section2\"\u003e \u003ch2\u003e3.2. SigMine: Software Adaptation\u003c/h2\u003e \u003cp\u003eThe workflow of SigMine was obtained through a sequence of automated steps, as illustrated in Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e and Supplementary video. The implementation provided a GUI for the different modules.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003e \u003cul\u003e \u003cli\u003e \u003cp\u003eThe Project manager module (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003ea and \u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eb) allows the user to create a new project or open existing projects. This ensures the formation or opening of a single system folder containing all the files related to the project.\u003c/p\u003e \u003c/li\u003e \u003cli\u003e \u003cp\u003eUpon launching the project, the left-hand side panel shows the entire Project Pipeline consisting of Welcome, PMCID Manager, Filtering, Data Cleaning, Analysis and Results.\u003c/p\u003e \u003c/li\u003e \u003cli\u003e \u003cp\u003eThe right-hand panel shows the main working interface, organized into six tabs:\u003c/p\u003e \u003c/li\u003e \u003c/ul\u003e \u003c/p\u003e \u003cp\u003e \u003col\u003e \u003cspan\u003e \u003cli\u003e \u003cp\u003eWelcome Tab: Confirmation of User ID/ Folder name (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003ec).\u003c/p\u003e \u003c/li\u003e \u003c/span\u003e \u003cspan\u003e \u003cli\u003e \u003cp\u003ePMCID Manager: Represents steps 2.1 and 2.2 of Methodology corresponding to input of PMCIDs and biomedical entity\u0026ndash;attribute recognition. PMCIDs can uploaded using the input box and then be saved, cleared, or submitted for annotation using the \u003cem\u003eSave\u003c/em\u003e, \u003cem\u003eClear\u003c/em\u003e and \u003cem\u003eSubmit\u003c/em\u003e buttons given below the input box as shown in Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003ed. The annotated output is saved in the Project folder as multiple csv formatted files, one corresponding to each type of annotation obtained e.g. organism, disease, tissues, etc.\u003c/p\u003e \u003c/li\u003e \u003c/span\u003e \u003cspan\u003e \u003cli\u003e \u003cp\u003eFiltering: Corresponds to steps 2.3 or data pre-processing and cleaning stage of the Methodology. It is performed in the Filtering tab (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003ee). Multiple annotation files from the project folder are loaded automatically using the \u003cem\u003eCheck Annotations\u003c/em\u003e button. An Annotation information window shows the total number of annotations obtained from file. It is processed using the \u003cem\u003eFilter Annotations\u003c/em\u003e button to remove null and redundant entries and the progress can be seen in the Filtering options window. A dialog box pos up and reports the process completion. The output is saved as multiple cleaned files in the project folder.\u003c/p\u003e \u003c/li\u003e \u003c/span\u003e \u003cspan\u003e \u003cli\u003e \u003cp\u003eData Cleaning: Corresponds to Step 2.4 described in the Methodology section or curated database-based standardization and is illustrated Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003ef. Multiple filtered annotation files are loaded using the \u003cem\u003eCheck Filtered Annotations\u003c/em\u003e button and standardized against curated databases using the \u003cem\u003eClean Data\u003c/em\u003e button. The entries uploaded can be seen in the Filtered annotation window while the progress can be monitored through the cleaning options window. The cleaned datasets are merged into a single dataframe using the \u003cem\u003eMerge\u003c/em\u003e button (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003ef). Popup dialogue boxes report successful completion at every stage. The output is saved as a single csv file.\u003c/p\u003e \u003c/li\u003e \u003c/span\u003e \u003cspan\u003e \u003cli\u003e \u003cp\u003eAnalysis: Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eg depicts the step 2.5 or Statistical evaluation stage of the methodology. The entity\u0026ndash;attribute pairs are analyzed using a configurable p-value threshold via the \u003cem\u003eRun Analysis\u003c/em\u003e button. The output is saved as multiple files containing each entity attribute pair and the corresponding p-value (lower than the cutoff specified).\u003c/p\u003e \u003c/li\u003e \u003c/span\u003e \u003cspan\u003e \u003cli\u003e \u003cp\u003eResults: Represents the step 2.6 or merging of significant entity-attribute pairs stage of the methodology (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eh) using the \u003cem\u003eCategorize \u0026amp; Merge Results\u003c/em\u003e button. The final output is then saved as a CSV file in the project directory.\u003c/p\u003e \u003c/li\u003e \u003c/span\u003e \u003c/ol\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec19\" class=\"Section2\"\u003e \u003ch2\u003e3.3. Comparison of SigMine with other methods\u003c/h2\u003e \u003cp\u003eMost existing literature-mining tools focus on specific tasks such as disease\u0026ndash;gene association (e.g., DTMiner), phenotype annotation (e.g., Ju et al., \u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e2019\u003c/span\u003e for COPD phenotype), or document classification (e.g., BioReader), and are often limited to abstracts or curated datasets (e.g., MeInfoText). The majority of these methods do not support multi-entity integration and lack statistical validation. A comparison of SigMine with other available methods is shown in Table\u0026nbsp;\u003cspan refid=\"Tab2\" class=\"InternalRef\"\u003e2\u003c/span\u003e. As evident from the Table, SigMine supports multiple biological entities, enables integration across different data types, and includes statistical analysis for identifying significant associations. It also provides a modular workflow that allows flexible data input and processing, making it more comprehensive in comparison with existing approaches.\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab2\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 2\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eComparison Of SigMine With Existing Literature-Mining Approaches\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"7\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c7\" colnum=\"7\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eTool / Study\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003ePrimary focus\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eInput scope\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eSupported entities\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003eMulti-entity integration\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c6\"\u003e \u003cp\u003eStatistical validation\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c7\"\u003e \u003cp\u003eAutomation level\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eGajendran et al., \u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e2007\u003c/span\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eDisease-gene discovery (bone biology)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003ePubMed abstracts\u0026thinsp;+\u0026thinsp;genomics\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eGenes, disease\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eNo\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003eNo\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003eSemi-automated\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eJu et al., \u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e2019\u003c/span\u003e (COPD phenotypes)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003ePhenotype annotation\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eCurated corpus\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eDisease phenotypes\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eLimited\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003eNo\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003eSemi-automated\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eLiu et al., \u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e2023\u003c/span\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003ePathway-centric knowledge graph\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eLiterature\u0026thinsp;+\u0026thinsp;pathway DBs\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eDrugs, microbes, neurotransmitters\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003ePartial\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003eNo\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003eSemi-automated\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eBioReader\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eDocument classification\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eAbstracts\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eText categories\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eNo\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003eNo\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003eAutomated\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eBioNerDS\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eTool/database usage trends\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eAbstracts\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eTools, databases\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eNo\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003eNo\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003eAutomated\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eOmixLitMiner\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eOmics lead prioritization\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eLiterature\u0026thinsp;+\u0026thinsp;omics data\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eGenes, proteins\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eNo\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003eNo\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003eSemi-automated\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eDTMiner\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eDisease-gene associations\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eAbstracts\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eDisease, gene\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eNo\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003eNo\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003eAutomated\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eMeInfoText\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eGene methylation in cancer\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eAbstracts\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eGenes, methylation\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eNo\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003eNo\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003eAutomated\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eTextpresso\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eOntology-based entity extraction\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eFull text\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eEntity mentions\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eNo\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003eNo\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003eSemi-automated\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eiTextMine\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eKnowledge graph construction\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eFull text\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eDisease, miRNA, drugs\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003ePartial\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003eNo\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003eAutomated\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePark et al., \u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e2021\u003c/span\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eMicrobe\u0026ndash;disease associations\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eAbstracts\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eMicrobe, disease\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eNo\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003eNo\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003eAutomated (DL-based)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eWu et al., \u003cspan citationid=\"CR65\" class=\"CitationRef\"\u003e2021\u003c/span\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eMicrobe\u0026ndash;disease associations\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eAbstracts\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eMicrobe, disease\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eNo\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003eNo\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003eAutomated (DL-based)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eEviMass\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eHuman microbial associations\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eAbstracts\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eMicrobe, disease\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eNo\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003eFisher test\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003eSemi-automated\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eSigMine (this study)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eGeneralized literature mining\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e\u003cb\u003eFull-text PMC\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e\u003cb\u003eOrganisms, diseases, tissues, genes/proteins, metabolites\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e\u003cb\u003eYes\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e\u003cb\u003eYes (χ\u0026sup2; + Bonferroni)\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e\u003cb\u003eFully\u003c/b\u003e\u003c/p\u003e \u003cp\u003e\u003cb\u003eautomated\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec20\" class=\"Section2\"\u003e \u003ch2\u003e3.4. Use case: Literature mining of human opportunistic pathogens\u003c/h2\u003e \u003cp\u003eThe pipeline was applied to mine human opportunistic pathogens from the PMC database, and uncover their biologically significant associations. The pipeline was implemented using the list of PMCIDs obtained with the keywords \u0026ldquo;human\u0026rdquo; AND \u0026ldquo;opportunistic pathogen\u0026rdquo;. As a result, a total of 25,425 research papers on human opportunistic pathogens were obtained from the NCBI PMC database. Using the PMCIDs obtained, annotations for organisms, diseases, genes, proteins, metabolites, and tissues were retrieved. Data cleaning and standardization were performed as described in the methodology to ensure high-quality data integration. Significant associations between opportunistic pathogens and their associated attributes were obtained through statistical analysis and are summarized in Table\u0026nbsp;\u003cspan refid=\"Tab3\" class=\"InternalRef\"\u003e3\u003c/span\u003e. Associated entity\u0026ndash;attribute pairs according to different levels of statistical significance were determined and are shown in Table\u0026nbsp;\u003cspan refid=\"Tab4\" class=\"InternalRef\"\u003e4\u003c/span\u003e. Clusters containing only highly significant associations represent the strongest relationships between opportunistic pathogens and their associated attributes, with gene/protein associations showing the highest representation.\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab3\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 3\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eCounts of the Statistically Significant Biomedical Entity-Attribute Associations\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"3\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\"\u0026times;\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eNumber of Opportunistic Pathogens\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eNumber and Types of Associated Attributes\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eCorrected P-value range\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e5626\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e1440\u003c/p\u003e \u003cp\u003eDisease\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\"\u0026times;\" colname=\"c3\"\u003e \u003cp\u003e3.87 \u0026times; 10⁻\u0026sup3;⁰⁰ \u0026ndash; 0.04996\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e227\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e30\u003c/p\u003e \u003cp\u003eTissue\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\"\u0026times;\" colname=\"c3\"\u003e \u003cp\u003e6.81 \u0026times; 10⁻⁶⁶ \u0026ndash; 0.04995\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e6788\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e7121\u003c/p\u003e \u003cp\u003eGene-Protein\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\"\u0026times;\" colname=\"c3\"\u003e \u003cp\u003e3.00 \u0026times; 10⁻\u0026sup3;⁰⁵ \u0026ndash; 0.04995\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e3204\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e598\u003c/p\u003e \u003cp\u003eMetabolite\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\"\u0026times;\" colname=\"c3\"\u003e \u003cp\u003e2.94 \u0026times; 10⁻\u0026sup2;⁹⁸ \u0026ndash; 0.04987\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab4\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 4\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eOrganism-Centric Clusters of Opportunistic Pathogens at Different Significance Levels\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"6\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eCluster Type\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eOpportunistic Pathogens\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eAssociated Diseases\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eAssociated Tissues\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003eAssociated Genes/Proteins\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c6\"\u003e \u003cp\u003eAssociated Metabolites\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAll highly significant entities\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e\u003cb\u003e5183\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e\u003cb\u003e1342\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e\u003cb\u003e25\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e\u003cb\u003e6391\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e\u003cb\u003e551\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e4 highly significant\u0026thinsp;+\u0026thinsp;1 moderately significant\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e\u003cb\u003e5183\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e\u003cb\u003e1342\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e\u003cb\u003e25\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e\u003cb\u003e6391\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e\u003cspan type=\"Underline\" class=\"Underline\" name=\"Emphasis\"\u003e231\u003c/span\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e3 highly significant\u0026thinsp;+\u0026thinsp;2 moderately significant\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e\u003cb\u003e5183\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e\u003cb\u003e1342\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e\u003cb\u003e25\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e\u003cspan type=\"Underline\" class=\"Underline\" name=\"Emphasis\"\u003e2335\u003c/span\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e\u003cspan type=\"Underline\" class=\"Underline\" name=\"Emphasis\"\u003e231\u003c/span\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e2 highly significant\u0026thinsp;+\u0026thinsp;3 moderately significant\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e\u003cb\u003e5183\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e\u003cb\u003e1342\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e\u003cspan type=\"Underline\" class=\"Underline\" name=\"Emphasis\"\u003e21\u003c/span\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e\u003cspan type=\"Underline\" class=\"Underline\" name=\"Emphasis\"\u003e2335\u003c/span\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e\u003cspan type=\"Underline\" class=\"Underline\" name=\"Emphasis\"\u003e231\u003c/span\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAll moderately significant entities\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e\u003cspan type=\"Underline\" class=\"Underline\" name=\"Emphasis\"\u003e1111\u003c/span\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e\u003cspan type=\"Underline\" class=\"Underline\" name=\"Emphasis\"\u003e640\u003c/span\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e\u003cspan type=\"Underline\" class=\"Underline\" name=\"Emphasis\"\u003e9\u003c/span\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e\u003cspan type=\"Underline\" class=\"Underline\" name=\"Emphasis\"\u003e1457\u003c/span\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e\u003cspan type=\"Underline\" class=\"Underline\" name=\"Emphasis\"\u003e105\u003c/span\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eOnly significant entities\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e1006\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e584\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e14\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e876\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e104\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003ctfoot\u003e \u003ctr\u003e\u003ctd colspan=\"6\"\u003eSignificance thresholds: highly significant (adjusted p-value\u0026thinsp;\u0026lt;\u0026thinsp;0.001), moderately significant (0.001\u0026thinsp;\u0026le;\u0026thinsp;adjusted p-value\u0026thinsp;\u0026lt;\u0026thinsp;0.01), significant (adjusted p-value\u0026thinsp;\u0026ge;\u0026thinsp;0.01).\u003c/td\u003e\u003c/tr\u003e \u003c/tfoot\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003eOPathDb provides a web interface for exploring significantly associated opportunistic pathogen\u0026ndash;based relationships identified through the biomedical literature mining pipeline, as illustrated in Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003e. The database contains information on opportunistic pathogens and their associated diseases, tissues, genes/proteins, and metabolites. The home page includes an overview of the database and a search form for querying the data (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003ea and \u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eb). The search module provides access to different entity categories, including organism, disease, tissue, gene/protein, and metabolite. This enables retrieval of opportunistic pathogen\u0026ndash;associated information. Keyword-based retrieval of entity-specific data is shown in the search interface, such as an organism name in the Organism category (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003ec). The retrieved results are displayed in a structured tabular format (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003ed).\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eOrganism-centric multipartite network visualizations representing significant pathogen\u0026ndash;attribute associations were obtained in OPathDb (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003e). This network depicts the pathogen as the central node, with its associated diseases, genes/proteins, metabolites, and tissues as connected nodes. Different node colors indicate different biological entity types, whereas edge colors and weights represent levels of statistical significance (3\u0026thinsp;=\u0026thinsp;high, 2\u0026thinsp;=\u0026thinsp;moderate, 1\u0026thinsp;=\u0026thinsp;significant). Additional information regarding database functionality and usage is available in the integrated FAQ/Tutorial section (Supplementary Table S2).\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec21\" class=\"Section2\"\u003e \u003ch2\u003e3.5. Comparison OpathDb data with other sources\u003c/h2\u003e \u003cp\u003eIn order to validate OPathDb against established pathogen data repositories, a pathogen-entry-based comparison was performed against the NHSN organism list (\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://www.cdc.gov/nhsn/about-nhsn/index.html\u003c/span\u003e\u003cspan address=\"https://www.cdc.gov/nhsn/about-nhsn/index.html\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e) and the gcPathogen database (Guo et al. \u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e2024\u003c/span\u003e). Using direct pathogen mapping, OpathDb matched 924 out of 2,278 NHSN-listed organisms (40.56% coverage) and 428 out of 858 gcPathogen organisms (49.88% coverage). Importantly, pathogens that do not overlap with these reference resources are literature-derived entries uniquely identified by the SigMine pipeline. This highlights the ability of SigMine to capture novel entity associations that are not explicitly catalogued in existing surveillance lists or pathogen databases.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec22\" class=\"Section2\"\u003e \u003ch2\u003e3.6. Case study-based validation of OpathDb\u003c/h2\u003e \u003cp\u003eThe correctness of the data in OPathDb was checked by examining the details of prominent categories of opportunistic pathogens as found in literature. Table\u0026nbsp;\u003cspan refid=\"Tab5\" class=\"InternalRef\"\u003e5\u003c/span\u003e summarizes representative opportunistic pathogens and their highly significant mined associations in OPathDb. Gastrointestinal taxa including \u003cem\u003eAkkermansia muciniphila\u003c/em\u003e (Zheng et al. \u003cspan citationid=\"CR77\" class=\"CitationRef\"\u003e2023\u003c/span\u003e; Gubernatorova et al. \u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e2023\u003c/span\u003e; Siwo et al. \u003cspan citationid=\"CR53\" class=\"CitationRef\"\u003e2024\u003c/span\u003e; Zhang et al. \u003cspan citationid=\"CR74\" class=\"CitationRef\"\u003e2025\u003c/span\u003e), \u003cem\u003eBacteroides fragilis\u003c/em\u003e (S\u0026aacute;nchez et al. \u003cspan citationid=\"CR45\" class=\"CitationRef\"\u003e2012\u003c/span\u003e), and \u003cem\u003eSegatella copri\u003c/em\u003e (Abdelsalam et al. \u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e2023\u003c/span\u003e; Xiao et al. \u003cspan citationid=\"CR68\" class=\"CitationRef\"\u003e2024b\u003c/span\u003e, \u003cspan citationid=\"CR67\" class=\"CitationRef\"\u003ea\u003c/span\u003e; Zhou et al. \u003cspan citationid=\"CR78\" class=\"CitationRef\"\u003e2025\u003c/span\u003e) were associated with colorectal cancer, celiac disease, and glucose intolerance, respectively, consistent with existing literature. Oral pathogens such as \u003cem\u003eFusobacterium nucleatum\u003c/em\u003e (Liu et al. \u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e2014\u003c/span\u003e; Dahlstrand Rudin et al. \u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e2021\u003c/span\u003e; Chen et al. \u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e2022a\u003c/span\u003e, \u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003eb\u003c/span\u003e; Wang et al. \u003cspan citationid=\"CR60\" class=\"CitationRef\"\u003e2023\u003c/span\u003e) and \u003cem\u003eTreponema denticola\u003c/em\u003e (Nakagawa et al. \u003cspan citationid=\"CR37\" class=\"CitationRef\"\u003e2002\u003c/span\u003e; Listyarifah et al. \u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e2018\u003c/span\u003e; Zang et al. \u003cspan citationid=\"CR73\" class=\"CitationRef\"\u003e2022\u003c/span\u003e) were linked to periodontal and oral cancer contexts, accompanied by known adhesins, stress-response proteins, and inflammatory metabolites as substantiated by previous studies. Cutaneous pathogens including \u003cem\u003eCutibacterium acne\u003c/em\u003e (Won et al. \u003cspan citationid=\"CR63\" class=\"CitationRef\"\u003e2022\u003c/span\u003e; Sauer et al. \u003cspan citationid=\"CR46\" class=\"CitationRef\"\u003e2024\u003c/span\u003e; Chen et al. \u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e2025a\u003c/span\u003e; Burma and Ramien \u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e2025\u003c/span\u003e), \u003cem\u003eEnterococcus faecium\u003c/em\u003e (Sharifi et al. \u003cspan citationid=\"CR51\" class=\"CitationRef\"\u003e2013\u003c/span\u003e; Goel et al. \u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e2016\u003c/span\u003e; Arshadi et al. \u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e2018\u003c/span\u003e; Codelia-Anjum et al. \u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e2023\u003c/span\u003e; Wu et al. \u003cspan citationid=\"CR66\" class=\"CitationRef\"\u003e2024\u003c/span\u003e), \u003cem\u003eKlebsiella pneumoniae\u003c/em\u003e (Wiskur et al. \u003cspan citationid=\"CR62\" class=\"CitationRef\"\u003e2008\u003c/span\u003e; Sridhar et al. \u003cspan citationid=\"CR55\" class=\"CitationRef\"\u003e2014\u003c/span\u003e; Serban et al. \u003cspan citationid=\"CR50\" class=\"CitationRef\"\u003e2021\u003c/span\u003e; Ren et al. \u003cspan citationid=\"CR42\" class=\"CitationRef\"\u003e2023\u003c/span\u003e), and \u003cem\u003ePseudomonas aeruginosa\u003c/em\u003e (Koh et al. \u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e2005\u003c/span\u003e; Woods et al. \u003cspan citationid=\"CR64\" class=\"CitationRef\"\u003e2011\u003c/span\u003e; Nesher et al. \u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e2015\u003c/span\u003e; Qin et al. \u003cspan citationid=\"CR41\" class=\"CitationRef\"\u003e2022\u003c/span\u003e) were associated with alopecia areata, urinary tract infection, endophthalmitis, and neutropenia, respectively, as stated in previous studies. Organisms uniquely present in OpathDb but not in the NHSN list and gcPathogen database included \u003cem\u003eA. muciniphila, S. copri\u003c/em\u003e and \u003cem\u003eT. denticola\u003c/em\u003e. Collectively, these examples indicate that OPathDb accurately consolidates literature-supported opportunistic pathogen\u0026ndash;attribute relationships throughout diverse tissues while capturing relevant molecular and metabolic signatures.\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab5\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 5\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eCase study\u0026ndash;based validation of highly significant opportunistic pathogen\u0026ndash;attributes associations captured in OpathDb\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"6\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eOpportunistic Pathogen\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eAssociated Disease\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eAssociated Tissue\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eAssociated Gene \u0026amp; Protein\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003eAssociated Metabolite\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c6\"\u003e \u003cp\u003eReferences\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cem\u003eAkkermansia muciniphila\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eColorectal Cancer\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eGastrointestinal Tract\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eTLR2, NLRP3, LPS, TLR4, MUC2, FMT\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eLactic acid\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e[50\u0026ndash;53]\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cem\u003eBacteroides fragilis\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eCeliac Disease\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eGastrointestinal Tract\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eTLR2, TLR4, NOD2, MUC2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eInulin\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e[54]\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cem\u003eCutibacterium acnes\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eAlopecia Areata\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eSkin\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eRECA, COL11A2, COL9A2, COL9A3\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eAzelaic acid\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e[67\u0026ndash;70]\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cem\u003eEnterococcus faecium\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eUrinary Tract Infection\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eUrinary System\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eAcm, Ebp, Ef1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eColistin, Tigecycline\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e[71\u0026ndash;75]\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cem\u003eFusobacterium nucleatum\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eAggressive Periontidis\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eMouth\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eCXCL9, CXCL10, CXCL16,CDH17, GAPDH, JNK\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eButyric acid\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e[59\u0026ndash;63]\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cem\u003eKlebsiella pneumoniae\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eEndophthalmitis\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eUrinary System\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eCps\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eAerobactin\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e[76\u0026ndash;79]\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cem\u003ePseudomonas aeruginosa\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eNeutropenia\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eFaeces\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eExsA, LasA, ToxA, PilA, OprF, AlgD, FliC, PopB, Rhl, Ly6G, CFTR\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e2-heptyl-3-hydroxy-4-quinolone\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e[80\u0026ndash;83]\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cem\u003eSegatella copri\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eGlucose Intolerance\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eFaeces\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eBSH, DCA, LBP\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003ePhenylacetylglutamine, Sedoheptulose 7-phosphate, Trimethylamine\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e[55\u0026ndash;58]\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cem\u003eTreponema denticola\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eTongue squamous cell carcinoma\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eMouth\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eKGP, LPS, FimA\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eButyric acid\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e[64\u0026ndash;66]\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003c/div\u003e"},{"header":"4. Conclusion","content":"\u003cp\u003eThis study presents SigMine, a scalable and adaptive literature-mining pipeline designed to systematically process large volumes of unstructured biomedical text in the context of structured biological knowledge to reliably elucidate significant associations among biological entities. The pipeline integrates entity recognition using machine-learning and deep learning\u0026ndash;based text mining along with other online-based resources, followed by statistical evaluation to identify significant associations. The framework is readily adaptable for generating domain-specific knowledgebases to support clinical research, surveillance, and evidence-driven decision-making. Thus, this study makes a meaningful contribution toward addressing significant literature mining challenges in computational biology by developing an automated scalable pipeline that: (a) efficiently handles large-scale entity annotation and association discovery, (b) integrates multiple bioinformatics tools, databases and standardized ontologies for robust curation, (c) statistically validates results for biological significance, d) implemented in a database, and e) provides modularity with intermittent input/output, enabling the user to upload data from any source to be cleaned, standardized, annotated, and subjected to merging, statistical tests, and multi-entity clustering.\u003c/p\u003e \u003cp\u003e. While remaining flexible for extension via dictionary and ontology updates, SigMine can be easily adapted for large-scale literature mining to address a wide range of associations such as host-pathogen, disease-disease, disease-gene, disease-metabolite, drug-pathway, drug-disease, drug-metabolite, drug\u0026ndash;target interactions, and pathway-level relationships.\u003c/p\u003e \u003cp\u003eSigMine was applied as a use case to mine human opportunistic pathogens and their associated diseases, tissues, genes, and metabolites. This resulted in the development of OPathDb, a curated database of pathogen-centric associations. The pipeline enables the generation of pathogen\u0026ndash;attribute clustered by statistical significance levels that can be viewed in tabular form or graphically in OPathDb. The current annotation framework focuses on organisms, diseases, tissues, genes/proteins, and metabolites but provides a flexible basis for further expansion. It can be extended to include additional biomedical entities and relationships. This enhances its application for large-scale knowledge discovery and supporting varied uses in biomedical and clinical research.\u003c/p\u003e"},{"header":"Abbreviations","content":"\u003cp\u003eAPI, Application Programming Interface; BTO, BRENDA Tissue Ontology; CSV, Comma-Separated Values; CSS, Cascading Style Sheet; FMA, Foundational Model of Anatomy; GUI, Graphical User Interface; HMDAD, Human Microbe-Disease Association Database; HMDB, Human Metabolome Database; \u0026nbsp;HTML, Hypertext Markup Language; MiMeDB, Microbial Metabolome Database; NCBI, National Center of Biotechnology Information; NLP, Natural Language Processing; PHP, Hypertext Preprocessor; PMC, PubMed Central\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eFunding:\u0026nbsp;\u003c/strong\u003eUR acknowledges research support from Netaji Subhas University of Technology, New Delhi, through a grant of University Research Fellowship. SB acknowledges support from DBT project BT/PR40197/BTIS/137/68/2023.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCompeting Interest:\u0026nbsp;\u003c/strong\u003eThe authors have no relevant financial or non-financial interests to disclose.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAuthor Contribution:\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eUrvija Rani:\u0026nbsp;\u003c/strong\u003eMethodology, Formal Analysis and Writing; \u003cstrong\u003eAkshath Nair:\u0026nbsp;\u003c/strong\u003eSoftware Implementation; \u003cstrong\u003eSonika Bhatnagar:\u0026nbsp;\u003c/strong\u003eConceptualization,Supervision, Project management and Review\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eORCiD Information\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eUrvija Rani: https://orcid.org/0000-0001-5520-8691\u003c/p\u003e\n\u003cp\u003eSonika Bhatnagar: https://orcid.org/0000-0002-8818-4240\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAcknowledgements:\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eWe gratefully acknowledge the funding and use of High Performance Computing Facility from NSUT, New Delhi.\u003c/p\u003e\u003ch2\u003eData Availability\u003c/h2\u003e\u003cp\u003eThe data generated is available in a the OPathDb database (https://www.opathdb.cbsblab-nsut.in).\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n \u003cli\u003eAbdelsalam NA, Hegazy SM, Aziz RK (2023) The curious case of Prevotella copri. Gut Microbes 15:. https://doi.org/10.1080/19490976.2023.2249152\u003c/li\u003e\n \u003cli\u003eArciola CR, Campoccia D, Montanaro L (2018) Implant infections: adhesion, biofilm formation and immune evasion. Nature Reviews Microbiology 2018 16:7 16:397\u0026ndash;409. https://doi.org/10.1038/s41579-018-0019-y\u003c/li\u003e\n \u003cli\u003eArshadi M, Mahmoudi M, Motahar MS, et al (2018) Virulence Determinants and Antimicrobial Resistance Patterns of Vancomycin-resistant Enterococcus faecium Isolated from Different Sources in Southwest Iran. Iran J Public Health 47:264\u003c/li\u003e\n \u003cli\u003eAurrecoechea C, Barreto A, Basenko EY, et al (2017) EuPathDB: the eukaryotic pathogen genomics database resource. Nucleic Acids Res 45:D581\u0026ndash;D591. https://doi.org/10.1093/NAR/GKW1105\u003c/li\u003e\n \u003cli\u003eBateman A, Martin MJ, Orchard S, et al (2025) UniProt: the Universal Protein Knowledgebase in 2025. Nucleic Acids Res 53:D609\u0026ndash;D617. https://doi.org/10.1093/NAR/GKAE1010\u003c/li\u003e\n \u003cli\u003eBurma NE, Ramien ML (2025) Cutaneous and Gut Dysbiosis in Alopecia Areata: A Review. JID Innov 5:100363. https://doi.org/10.1016/j.xjidi.2025.100363\u003c/li\u003e\n \u003cli\u003eCasadevall A, Pirofski LA (2009) Virulence factors and their mechanisms of action: the view from a damage\u0026ndash;response framework. J Water Health 7:S2\u0026ndash;S18. https://doi.org/10.2166/WH.2009.036\u003c/li\u003e\n \u003cli\u003eChebotar I, Azizov I, Edelstein M, Kozlov R (2025) Detection of virulence factors in opportunistic bacteria: advances, challenges, and practical implementation. Front Microbiol 16:1638925. https://doi.org/10.3389/FMICB.2025.1638925/FULL\u003c/li\u003e\n \u003cli\u003eChen Q, Liu C, Tao J, et al (2025a) Insights into microbial dysbiosis and Cutibacterium acnes CAMP factor interactions in acne vulgaris. Microb Genom 11:001449. https://doi.org/10.1099/mgen.0.001449\u003c/li\u003e\n \u003cli\u003eChen S, Liao L, Wang M (2025b) Editorial: Opportunistic pathogens: pathogenesis and multi-drug resistance mechanisms. Front Microbiol 16:1597769. https://doi.org/10.3389/FMICB.2025.1597769/BIBTEX\u003c/li\u003e\n \u003cli\u003eChen Y, Huang Z, Tang Z, et al (2022a) More Than Just a Periodontal Pathogen \u0026ndash;the Research Progress on Fusobacterium nucleatum. Front Cell Infect Microbiol 12:815318. https://doi.org/10.3389/fcimb.2022.815318\u003c/li\u003e\n \u003cli\u003eChen Y, Shi T, Li Y, et al (2022b) Fusobacterium nucleatum: The Opportunistic Pathogen of Periodontal and Peri-Implant Diseases. Front Microbiol 13:860149. https://doi.org/10.3389/FMICB.2022.860149/XML\u003c/li\u003e\n \u003cli\u003eCodelia-Anjum A, Lerner LB, Elterman D, et al (2023) Enterococcal Urinary Tract Infections: A Review of the Pathogenicity, Epidemiology, and Treatment. Antibiotics 12:778. https://doi.org/10.3390/antibiotics12040778\u003c/li\u003e\n \u003cli\u003eDahlstrand Rudin A, Khamzeh A, Venkatakrishnan V, et al (2021) Short chain fatty acids released by Fusobacterium nucleatum are neutrophil chemoattractants acting via free fatty acid receptor 2 (FFAR2). Cell Microbiol 23:e13348. https://doi.org/10.1111/cmi.13348\u003c/li\u003e\n \u003cli\u003eDuck G, Nenadic G, Brass A, et al (2013) BioNerDS: Exploring bioinformatics\u0026rsquo; database and software use through literature mining. BMC Bioinformatics 14:1\u0026ndash;13. https://doi.org/10.1186/1471-2105-14-194/FIGURES/7\u003c/li\u003e\n \u003cli\u003eEspe S (2018) Malacards: The Human Disease Database. J Med Libr Assoc 106:140. https://doi.org/10.5195/JMLA.2018.253\u003c/li\u003e\n \u003cli\u003eFang YC, Huang HC, Juan HF (2008) MeInfoText: associated gene methylation and cancer information from text mining. BMC Bioinformatics 9:22. https://doi.org/10.1186/1471-2105-9-22\u003c/li\u003e\n \u003cli\u003eFusco V, Abriouel H, Benomar N, et al (2018) Opportunistic Food-Borne Pathogens. Food Safety and Preservation 269\u0026ndash;306. https://doi.org/10.1016/B978-0-12-814956-0.00010-X\u003c/li\u003e\n \u003cli\u003eGajendran VK, Lin JR, Fyhrie DP (2007) An application of bioinformatics and text mining to the discovery of novel genes related to bone biology. Bone 40:1378\u0026ndash;1388. https://doi.org/10.1016/J.BONE.2006.12.067\u003c/li\u003e\n \u003cli\u003eGoel V, Kumar D, Kumar R, et al (2016) Community Acquired Enterococcal Urinary Tract Infections and Antibiotic Resistance Profile in North India. J Lab Physicians 8:50. https://doi.org/10.4103/0974-2727.176237\u003c/li\u003e\n \u003cli\u003eGremse M, Chang A, Schomburg I, et al (2011) The BRENDA Tissue Ontology (BTO): the first all-integrating ontology of all organisms for enzyme sources. Nucleic Acids Res 39:D507\u0026ndash;D513. https://doi.org/10.1093/NAR/GKQ968\u003c/li\u003e\n \u003cli\u003eGubernatorova EO, Gorshkova EA, Bondareva MA, et al (2023) Akkermansia muciniphila - friend or foe in colorectal cancer? Front Immunol 14:1303795. https://doi.org/10.3389/FIMMU.2023.1303795/BIBTEX\u003c/li\u003e\n \u003cli\u003eGuo C, Chen Q, Fan G, et al (2024) gcPathogen: a comprehensive genomic resource of human pathogens for public health. Nucleic Acids Res 52:D714\u0026ndash;D723. https://doi.org/10.1093/NAR/GKAD875\u003c/li\u003e\n \u003cli\u003eHopkins HA, Lopezguerra C, Lau MJ, Raymann K (2024) Making a Pathogen? Evaluating the Impact of Protist Predation on the Evolution of Virulence in Serratia marcescens. Genome Biol Evol 16:. https://doi.org/10.1093/GBE/EVAE149\u003c/li\u003e\n \u003cli\u003eHornef M (2015) Pathogens, Commensal Symbionts, and Pathobionts: Discovery and Functional Effects on the Host. ILAR J 56:159\u0026ndash;162. https://doi.org/10.1093/ILAR/ILV007\u003c/li\u003e\n \u003cli\u003eHtun ZM, Gul MH, Sadikot RT (2024) Bacterial Infections in Patients Living with HIV. Results Probl Cell Differ 73:537. https://doi.org/10.1007/978-3-031-62036-2_21\u003c/li\u003e\n \u003cli\u003eJanssens Y, Nielandt J, Bronselaer A, et al (2018) Disbiome database: Linking the microbiome to disease. BMC Microbiol 18:1\u0026ndash;6. https://doi.org/10.1186/S12866-018-1197-5/TABLES/1\u003c/li\u003e\n \u003cli\u003eJos\u0026eacute; RJ, Periselneris JN, Brown JS (2020) Opportunistic bacterial, viral and fungal infections of the lung. Medicine 48:366\u0026ndash;372. https://doi.org/10.1016/J.MPMED.2020.03.006\u003c/li\u003e\n \u003cli\u003eJu M, Short AD, Thompson P, et al (2019) Annotating and detecting phenotypic information for chronic obstructive pulmonary disease. JAMIA Open 2:261. https://doi.org/10.1093/JAMIAOPEN/OOZ009\u003c/li\u003e\n \u003cli\u003eKhatib Sulaiman Dalam No J, Atam Mustafa A, Mohsin Abdulazeez A (2024) A Review of Text Classification Based on ML \u0026amp; Data Mining Algorithms. The Indonesian Journal of Computer Science 13:. https://doi.org/10.33022/IJCS.V13I3.4027\u003c/li\u003e\n \u003cli\u003eKoh AY, Priebe GP, Pier GB (2005) Virulence of Pseudomonas aeruginosa in a Murine Model of Gastrointestinal Colonization and Dissemination in Neutropenia. Infect Immun 73:2262. https://doi.org/10.1128/IAI.73.4.2262-2272.2005\u003c/li\u003e\n \u003cli\u003eListyarifah D, Nieminen MT, M\u0026auml;kinen LK, et al (2018) Treponema denticola chymotrypsin-like proteinase is present in early-stage mobile tongue squamous cell carcinoma and related to the clinicopathological features. Journal of Oral Pathology and Medicine 47:764\u0026ndash;772. https://doi.org/10.1111/jop.12729\u003c/li\u003e\n \u003cli\u003eLiu P, Liu Y, Wang J, et al (2014) Detection of Fusobacterium Nucleatum and fadA Adhesin Gene in Patients with Orthodontic Gingivitis and Non-Orthodontic Periodontal Inflammation. PLoS One 9:e85280. https://doi.org/10.1371/journal.pone.0085280\u003c/li\u003e\n \u003cli\u003eLiu T, Feenstra KA, Huang Z, Heringa J (2023) Mining literature and pathway data to explore the relations of ketamine with neurotransmitters and gut microbiota using a knowledge-graph. Bioinformatics 40:btad771. https://doi.org/10.1093/BIOINFORMATICS/BTAD771\u003c/li\u003e\n \u003cli\u003eMa W, Zhang L, Zeng P, et al (2017) An analysis of human microbe\u0026ndash;disease associations. Brief Bioinform 18:85\u0026ndash;97. https://doi.org/10.1093/BIB/BBW005\u003c/li\u003e\n \u003cli\u003eM\u0026uuml;ller HM, Kenny EE, Sternberg PW (2004) Textpresso: An Ontology-Based Information Retrieval and Extraction System for Biological Literature. PLoS Biol 2:e309. https://doi.org/10.1371/JOURNAL.PBIO.0020309\u003c/li\u003e\n \u003cli\u003eNakagawa I, Amano A, Kuboniwa M, et al (2002) Functional Differences among FimA Variants of Porphyromonas gingivalis and Their Effects on Adhesion to and Invasion of Human Epithelial Cells. Infect Immun 70:277. https://doi.org/10.1128/IAI.70.1.277-285.2002\u003c/li\u003e\n \u003cli\u003eNesher L, Rolston KVI, Shah DP, et al (2015) Fecal colonization and infection with Pseudomonas aeruginosa in recipients of allogeneic hematopoietic stem cell transplantation. Transplant Infectious Disease 17:33\u0026ndash;38. https://doi.org/10.1111/tid.12323\u003c/li\u003e\n \u003cli\u003ePark Y, Lee J, Moon H, et al (2021) Discovering microbe-disease associations from the literature using a hierarchical long short-term memory network and an ensemble parser model. Scientific Reports 2021 11:1 11:1\u0026ndash;12. https://doi.org/10.1038/s41598-021-83966-8\u003c/li\u003e\n \u003cli\u003ePolvi EJ, Li X, O\u0026rsquo;Meara TR, et al (2015) Opportunistic yeast pathogens: reservoirs, virulence mechanisms, and therapeutic strategies. Cell Mol Life Sci 72:2261. https://doi.org/10.1007/S00018-015-1860-Z\u003c/li\u003e\n \u003cli\u003eQin S, Xiao W, Zhou C, et al (2022) Pseudomonas aeruginosa: pathogenesis, virulence factors, antibiotic resistance, interaction with host, technology advances and emerging therapeutics. Signal Transduction and Targeted Therapy 2022 7:1 7:1\u0026ndash;27. https://doi.org/10.1038/s41392-022-01056-1\u003c/li\u003e\n \u003cli\u003eRen C, Li Z, Meng F, et al (2023) Endogenous endophthalmitis caused by urinary tract infection: A case report. Medicine (United States) 102:E36139. https://doi.org/10.1097/MD.0000000000036139\u003c/li\u003e\n \u003cli\u003eRen J, Li G, Ross K, et al (2018) iTextMine: integrated text-mining system for large-scale knowledge extraction from the literature. Database (Oxford) 2018:bay128. https://doi.org/10.1093/DATABASE/BAY128\u003c/li\u003e\n \u003cli\u003eRosse C, Mejino JLV (2003) A reference ontology for biomedical informatics: the Foundational Model of Anatomy. J Biomed Inform 36:478\u0026ndash;500. https://doi.org/10.1016/J.JBI.2003.11.007\u003c/li\u003e\n \u003cli\u003eS\u0026aacute;nchez E, Laparra JM, Sanz Y (2012) Discerning the Role of Bacteroides fragilis in Celiac Disease Pathogenesis. Appl Environ Microbiol 78:6507. https://doi.org/10.1128/AEM.00563-12\u003c/li\u003e\n \u003cli\u003eSauer N, Oślizło M, Brzostek M, et al (2024) The multiple uses of azelaic acid in dermatology: mechanism of action, preparations, and potential therapeutic applications. Advances in Dermatology and Allergology/Postȩpy Dermatologii i Alergologii 40:716. https://doi.org/10.5114/ada.2023.133955\u003c/li\u003e\n \u003cli\u003eSayers E (2022) A General Introduction to the E-utilities\u003c/li\u003e\n \u003cli\u003eSchober I, Koblitz J, Carbasse JS, et al (2025) BacDive in 2025: the core database for prokaryotic strain data. Nucleic Acids Res 53:D748\u0026ndash;D756. https://doi.org/10.1093/NAR/GKAE959\u003c/li\u003e\n \u003cli\u003eSchoch CL, Ciufo S, Domrachev M, et al (2020) NCBI Taxonomy: a comprehensive update on curation, resources and tools. Database (Oxford) 2020:baaa062. https://doi.org/10.1093/DATABASE/BAAA062\u003c/li\u003e\n \u003cli\u003eSerban D, Popa Cherecheanu A, Dascalu AM, et al (2021) Hypervirulent Klebsiella pneumoniae Endogenous Endophthalmitis\u0026mdash;A Global Emerging Disease. 11:676. https://doi.org/10.3390/life11070676\u003c/li\u003e\n \u003cli\u003eSharifi Y, Hasani A, Ghotaslou R, et al (2013) Virulence and Antimicrobial Resistance in Enterococci Isolated from Urinary Tract Infections. Adv Pharm Bull 3:197. https://doi.org/10.5681/apb.2013.032\u003c/li\u003e\n \u003cli\u003eSimon C, Davidsen K, Hansen C, et al (2019) BioReader: A text mining tool for performing classification of biomedical literature. BMC Bioinformatics 19:165\u0026ndash;170. https://doi.org/10.1186/S12859-019-2607-X/TABLES/1\u003c/li\u003e\n \u003cli\u003eSiwo GH, Singal AG, Waljee AK (2024) Pan-cancer molecular signatures connecting aspartate transaminase (AST) to cancer prognosis, metabolic and immune signatures. bioRxiv 2024.03.01.582939. https://doi.org/10.1101/2024.03.01.582939\u003c/li\u003e\n \u003cli\u003eSosa EJ, Burguener G, Lanzarotti E, et al (2017) Target-Pathogen: a structural bioinformatic approach to prioritize drug targets in pathogens. Nucleic Acids Res 46:D413. https://doi.org/10.1093/NAR/GKX1015\u003c/li\u003e\n \u003cli\u003eSridhar J, Flynn HW, Kuriyan AE, et al (2014) Endophthalmitis caused by Klebsiella species. Retina 34:1875. https://doi.org/10.1097/IAE.0000000000000162\u003c/li\u003e\n \u003cli\u003eSrivastava D, Baksi KD, Kuntal BK, Mande SS (2019) \u0026ldquo;EviMass\u0026rdquo;: A Literature Evidence-Based Miner for Human Microbial Associations. Front Genet 10:466029. https://doi.org/10.3389/FGENE.2019.00849/BIBTEX\u003c/li\u003e\n \u003cli\u003eSteffen P, Wu J, Hariharan S, et al (2020) OmixLitMiner: A Bioinformatics Tool for Prioritizing Biological Leads from \u0026lsquo;Omics Data Using Literature Retrieval and Data Mining. International Journal of Molecular Sciences 2020, Vol 21, Page 1374 21:1374. https://doi.org/10.3390/IJMS21041374\u003c/li\u003e\n \u003cli\u003eStringer AM, Hargreaves BM, Mendes RA, et al (2024) Updated perspectives on the contribution of the microbiome to the pathogenesis of mucositis using the MASCC/ISOO framework. Supportive Care in Cancer 32:558. https://doi.org/10.1007/S00520-024-08752-4\u003c/li\u003e\n \u003cli\u003eVuotto C, Longo F, Balice MP, et al (2014) Antibiotic Resistance Related to Biofilm Formation in \u0026nbsp;Klebsiella pneumoniae. Pathogens 2014, Vol 3, Pages 743-758 3:743\u0026ndash;758. https://doi.org/10.3390/PATHOGENS3030743\u003c/li\u003e\n \u003cli\u003eWang Y, Wang L, Sun T, et al (2023) Study of the inflammatory activating process in the early stage of Fusobacterium nucleatum infected PDLSCs. International Journal of Oral Science 2023 15:1 15:8-. https://doi.org/10.1038/s41368-022-00213-0\u003c/li\u003e\n \u003cli\u003eWishart DS, Oler E, Peters H, et al (2023) MiMeDB: the Human Microbial Metabolome Database. Nucleic Acids Res 51:D611\u0026ndash;D620. https://doi.org/10.1093/NAR/GKAC868\u003c/li\u003e\n \u003cli\u003eWiskur BJ, Hunt JJ, Callegan MC (2008) Hypermucoviscosity as a Virulence Factor in Experimental Klebsiella pneumoniae Endophthalmitis. Invest Ophthalmol Vis Sci 49:4931. https://doi.org/10.1167/iovs.08-2276\u003c/li\u003e\n \u003cli\u003eWon EJ, Jang HH, Park H, Kim SJ (2022) A Potential Predictive Role of the Scalp Microbiome Profiling in Patients with Alopecia Areata: Staphylococcus caprae, Corynebacterium, and Cutibacterium Species. Microorganisms 10:864. https://doi.org/10.3390/microorganisms10050864\u003c/li\u003e\n \u003cli\u003eWoods DE, Lam JS, Paranchych W, et al (2011) Correlation of Pseudomonas aeruginosa virulence factors from clinical and environmental isolates with pathogenicity in the neutropenic mouse. https://doi.org/101139/m97-077 43:541\u0026ndash;551. https://doi.org/10.1139/m97-077\u003c/li\u003e\n \u003cli\u003eWu C, Xiao X, Yang C, et al (2021) Mining microbe\u0026ndash;disease interactions from literature via a transfer learning model. BMC Bioinformatics 22:1\u0026ndash;15. https://doi.org/10.1186/S12859-021-04346-7/TABLES/3\u003c/li\u003e\n \u003cli\u003eWu W, Xiao S, Han L, Wu Q (2024) Antimicrobial resistance, virulence gene profiles, and molecular epidemiology of enterococcal isolates from patients with urinary tract infections in Shanghai, China. Microbiol Spectr 13:e01217-24. https://doi.org/10.1128/spectrum.01217-24\u003c/li\u003e\n \u003cli\u003eXiao X, Singh A, Giometto A, Brito IL (2024a) Segatella clades adopt distinct roles within a single individual\u0026rsquo;s gut. NPJ Biofilms Microbiomes 10:114. https://doi.org/10.1038/s41522-024-00590-w\u003c/li\u003e\n \u003cli\u003eXiao X, Singh A, Giometto A, Brito IL (2024b) Segatella copri strains adopt distinct roles within a single individual\u0026rsquo;s gut. bioRxiv 2024.05.20.595015. https://doi.org/10.1101/2024.05.20.595015\u003c/li\u003e\n \u003cli\u003eXu D, Zhang M, Xie Y, et al (2016) DTMiner: identification of potential disease targets through biomedical literature mining. Bioinformatics 32:3619\u0026ndash;3626. https://doi.org/10.1093/BIOINFORMATICS/BTW503\u003c/li\u003e\n \u003cli\u003eYang X, Saha S, Venkatesan A, et al (2023) Europe PMC annotated full-text corpus for gene/proteins, diseases and organisms. Scientific Data 2023 10:1 10:1\u0026ndash;13. https://doi.org/10.1038/s41597-023-02617-x\u003c/li\u003e\n \u003cli\u003eYao G, Zhang W, Yang M, et al (2021) MicroPhenoDB Associates Metagenomic Data with Pathogenic Microbes, Microbial Core Genes, and Human Disease Phenotypes. Genomics Proteomics Bioinformatics 18:760. https://doi.org/10.1016/J.GPB.2020.11.001\u003c/li\u003e\n \u003cli\u003eYoungs J, Arnold A (2021) Infections in the immunocompromised host: secondary immunodeficiency disorders. Medicine 49:611\u0026ndash;617. https://doi.org/10.1016/J.MPMED.2021.07.003\u003c/li\u003e\n \u003cli\u003eZang W, Liu J, Geng F, et al (2022) Butyrate promotes oral squamous cell carcinoma cells migration, invasion and epithelial-mesenchymal transition. PeerJ 10:e12991. https://doi.org/10.7717/peerj.12991\u003c/li\u003e\n \u003cli\u003eZhang K, Zhu YW, Tang AQ, et al (2025) Role of 3-mercaptopyruvate sulfurtransferase in cancer: Molecular mechanisms and therapeutic perspectives. Transl Oncol 52:102272. https://doi.org/10.1016/j.tranon.2025.102272\u003c/li\u003e\n \u003cli\u003eZhang T, Miao J, Han N, et al (2018) MPD: a pathogen genome and metagenome database. Database 2018:. https://doi.org/10.1093/DATABASE/BAY055\u003c/li\u003e\n \u003cli\u003eZhao S, Su C, Lu Z, Wang F (2020) Recent advances in biomedical literature mining. Brief Bioinform 22:bbaa057. https://doi.org/10.1093/BIB/BBAA057\u003c/li\u003e\n \u003cli\u003eZheng M, Han R, Yuan Y, et al (2023) The role of Akkermansia muciniphila in inflammatory bowel disease: Current knowledge and perspectives. Front Immunol 13:1089600. https://doi.org/10.3389/FIMMU.2022.1089600/BIBTEX\u003c/li\u003e\n \u003cli\u003eZhou C, Hibberd MC, Lee EM, et al (2025) Glycoside hydrolase\u0026ndash;mediated glucomannan catabolism in Segatella copri, a target of microbiota-directed foods for malnourished children. Proc Natl Acad Sci U S A 122:e2521522122. https://doi.org/10.1073/pnas.2521522122\u003c/li\u003e\n\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"archives-of-microbiology","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"aomi","sideBox":"Learn more about [Archives of Microbiology](https://www.springer.com/journal/203)","snPcode":"203","submissionUrl":"https://submission.nature.com/new-submission/203/3","title":"Archives of Microbiology","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"em","reportingPortfolio":"Springer Hybrid","inReviewEnabled":true,"inReviewRevisionsEnabled":false},"keywords":"biomedical literature mining, opportunistic pathogens, entity, attribute, significance associations, database","lastPublishedDoi":"10.21203/rs.3.rs-9315983/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-9315983/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eRecognition of cross-domain association between biological entities in the vast biomedical literature is a challenging task. SigMine, an automated pipeline, was constructed to systematically mine biomedical literature to identify significantly associated biological entities. SigMine performs biomedical entity recognition from PMC articles using machine-learning and deep-learning\u0026ndash;based entity recognition through Europe PMC Annotation API. Advanced entity recognition using Python scripting, NCBI E-Utilities, and an n-gram algorithm was performed followed by extensive data cleaning and mapping against standard databases. Statistical evaluation identified significant associations between entities. The entire workflow was automated through a modular framework developed in Python v3.13 with a Tkinter-based Graphical User Interface. SigMine enhances usability while retaining the flexibility to use new dictionaries for annotation. SigMine was used to construct a human Opportunistic Pathogens Database (OPathDb), housing 5,626 novel opportunistic pathogens significantly associated with 1,440 diseases and 7,121 genes mined from 25,000 PMC articles. Additional annotation of 598 significantly associated metabolites and 30 affected tissues is available for 3,204 and 227 pathogens respectively. OpathDb has a user-friendly query interface searchable by organism, disease, tissue, gene, protein and metabolite available at \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://www.opathdb.cbsblab-nsut.in\u003c/span\u003e\u003cspan address=\"https://www.opathdb.cbsblab-nsut.in\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e. Organism\u0026ndash;entity associations can be visualized as weighted networks, with color-coded nodes and significance-scaled edges. Significant associations of opportunistic pathogens like \u003cem\u003eAkkermansia mucinifila\u003c/em\u003e with colorectal cancer and \u003cem\u003eSegatella copri\u003c/em\u003e with glucose intolerance can be identified through OpathDb. The SigMine framework demonstrates efficient recognition and prioritization of relationships in a vast and heterogenous corpora.\u003c/p\u003e","manuscriptTitle":"SigMine and OPathDb: A Literature-Mining Pipeline and Opportunistic Pathogen Database","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2026-04-13 17:10:28","doi":"10.21203/rs.3.rs-9315983/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"decision","content":"Revision requested","date":"2026-04-20T02:32:05+00:00","index":"","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2026-04-19T18:22:31+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2026-04-16T05:21:55+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2026-04-15T08:56:28+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2026-04-14T14:32:07+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"269203634208125140889203792142021540559","date":"2026-04-12T07:23:23+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2026-04-11T04:33:08+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"264697286262075606729302789555580910481","date":"2026-04-09T08:45:35+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"303990696204314455450742319417421034827","date":"2026-04-08T14:00:01+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"271030969410347991930020926926805206721","date":"2026-04-07T13:42:03+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"27444692213562473541705062110530970373","date":"2026-04-07T13:27:55+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"209695668991716138228889035375889434824","date":"2026-04-07T08:04:57+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"260628310336115458336602132627842410962","date":"2026-04-07T07:43:19+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2026-04-07T07:17:22+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2026-04-07T07:10:08+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2026-04-07T02:56:20+00:00","index":"","fulltext":""},{"type":"submitted","content":"Archives of Microbiology","date":"2026-04-03T20:04:20+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"archives-of-microbiology","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"aomi","sideBox":"Learn more about [Archives of Microbiology](https://www.springer.com/journal/203)","snPcode":"203","submissionUrl":"https://submission.nature.com/new-submission/203/3","title":"Archives of Microbiology","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"em","reportingPortfolio":"Springer Hybrid","inReviewEnabled":true,"inReviewRevisionsEnabled":false}}],"origin":"","ownerIdentity":"25b49e0f-232a-4483-918f-32d6ea69cf23","owner":[],"postedDate":"April 13th, 2026","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"in-revision","subjectAreas":[],"tags":[],"updatedAt":"2026-04-20T02:39:00+00:00","versionOfRecord":[],"versionCreatedAt":"2026-04-13 17:10:28","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-9315983","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-9315983","identity":"rs-9315983","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}