Post-Acute Infection Syndromes (PAIS) are medical conditions that persist following acute infections from pathogens such as SARS-CoV-2, Epstein-Barr virus, and Influenza virus. Despite growing global awareness of PAIS and the exponential increase in biomedical literature, only a small fraction of this literature pertains specifically to PAIS, making the identification of pathogen-disease associations within such a vast, heterogeneous, and unstructured corpus a significant challenge for researchers. This study evaluated the effectiveness of large language models (LLMs) in extracting these associations through a binary classification task using a curated dataset of 1000 manually labeled PubMed abstracts. We benchmarked a wide range of open-source LLMs of varying sizes (4B-70B parameters), including generalist, reasoning, and biomedical-specific models. We also investigated the extent to which prompting strategies such as zero-shot, few-shot, and Chain of Thought (CoT) methods can improve classification performance. Our results indicate that model performance varied by size, architecture, and prompting strategy. Zero-shot prompting produced the most reliable results: Mistral-Small-Instruct-2409 and Llama-3.1-Nemotron-70B-Instruct achieved balanced accuracy scores of 0.81 and 0.80, respectively, along with macro-F1 scores of up to 0.80, while maintaining minimal invalid outputs. While few-shot and CoT prompting often degraded performance in generalist models, reasoning models such as DeepSeek-R1-Distill-Llama-70B and QwQ-32B demonstrated improved accuracy and consistency when provided with additional context.

SUMMARY: Visualization of multidimensional, categorical data is a common challenge across scientific domains and, in particular, the life sciences. The goal is to create a comprehensive overview of the underlying data which enables one to assess multiple variables. One application where such visualizations are particularly useful is gene or pathway analysis, which involves checking for dysregulation in known biological mechanisms and functions across multiple conditions. Here, we propose a new visualization approach that encodes such data in an intuitive representation: DicePlots visualize up to four distinct categorical classes in a single view using elements resembling dice faces, whereas DominoPlots add an additional layer of information for binary comparison. AVAILABILITY AND IMPLEMENTATION: The code is available as the diceplot R package and the pydiceplot on PyPI. All source code is available at https://github.com/maflot. CONTACT: The repo is managed actively and we encourage community contributions and requests.

Estimating the depth of virtual content has proven to be a challenging task in Augmented Reality (AR) applications. Existing studies have shown that the visual system makes use of multiple depth cues to infer the distance of objects, occlusion being one of the most important ones. The ability to generate appropriate occlusions becomes particularly important for AR applications that require the visualization of augmented objects placed below a real surface. Examples of these applications are medical scenarios in which the visualization of anatomical information needs to be observed within the patient's body. In this regard, existing works have proposed several focus and context (F+C) approaches to aid users in visualizing this content using Video See-Through (VST) Head-Mounted Displays (HMDs). However, the implementation of these approaches in Optical See-Through (OST) HMDs remains an open question due to the additive characteristics of the display technology. In this article, we, for the first time, design and conduct a user study that compares depth estimation between VST and OST HMDs using existing in-situ visualization methods. Our results show that these visualizations cannot be directly transferred to OST displays without increasing error in depth perception tasks. To tackle this gap, we perform a structured decomposition of the visual properties of AR F+C methods to find best-performing combinations. We propose the use of chromatic shadows and hatching approaches transferred from computer graphics. In a second study, we perform a factorized analysis of these combinations, showing that varying the shading type and using colored shadows can lead to better depth estimation when using OST HMDs.

RNA sequencing data sets rapidly increase in quantity. For microRNAs (miRNAs), frequently dozens to hundreds of billion reads are generated per study. The quantification of annotated miRNAs and the prediction of new miRNAs are leading computational tasks. Now, the increased depth of coverage allows to gain deeper insights into the variability of miRNAs. The analysis of isoforms of miRNAs (isomiRs) is a trending topic, and a range of computational tools for the analysis of isomiRs has been developed. We provide an overview on 27 available computational solutions for the analysis of isomiRs. These include both stand-alone programs (17 tools) and web-based solutions (10 tools) and span a publication time range from 2010 to 2020. Seven of the tools were published in 2019 and 2020, confirming the rising importance of the topic. While most of the analyzed tools work for a broad range of organisms or are completely independent of a reference organism, several tools have been tailored for the analysis of human miRNA data or for plants. While 14 of the tools are general analysis tools of miRNAs, and isomiR analysis is one of their features, the remaining 13 tools have specifically been developed for isomiR analysis. A direct comparison on 20 deep sequencing data sets for selected tools provides insights into the heterogeneity of results. With our work, we provide users a comprehensive overview on the landscape of isomiR analysis tools and in that support the selection of the most appropriate tool for their respective research task.

MOTIVATION: Since the initial discovery of microRNAs as post-transcriptional, regulatory key players in the 1990s, a total number of $2656$ mature microRNAs have been publicly described for Homo sapiens. As discovery of new miRNAs is still on-going, target identification remains to be an essential and challenging step preceding functional annotation analysis. One key challenge for researchers seems to be the selection of the most appropriate tool out of the larger multiverse of published solutions for a given research study set-up. RESULTS: In this review we collectively describe the field of in silico target prediction in the course of time and point out long withstanding principles as well as recent developments. By compiling a catalog of characteristics about the 98 prediction methods and identifying common and exclusive traits, we signpost a simplified mechanism to address the problem of application selection. Going further we devised interpretation strategies for common types of output as generated by frequently used computational methods. To this end, our work specifically aims to make prospective users aware of common mistakes and practical questions that arise during the application of target prediction tools. AVAILABILITY: An interactive implementation of our recommendations including materials shown in the manuscript is freely available at https://www.ccb.uni-saarland.de/mtguide.

Modern precision medicine comprises the knowledge and understanding of individual differences in the genomic sequence of patients to provide tailor-made treatments. Regularly, such variants are considered in coding regions only, and their effects are predicted based on their impact on the amino acid sequence of expressed proteins. However, assessing the effects of variants in noncoding elements, in particular microRNAs (miRNAs) and their binding sites, is important as well, as a single miRNA can influence the expression patterns of many genes at the same time. To analyze the effects of variants in miRNAs and their target sites, several databases storing variant impact predictions have been published. In this review, we will compare the core functionalities and features of these databases and discuss the importance of up-to-date data resources in the context of web applications. Finally, we will outline some recommendations for future developments in the field.

High-throughput next-generation shotgun sequencing of pathogenic bacteria is growing in clinical relevance, especially for chromosomal DNA-based taxonomic identification and for antibiotic resistance prediction. Genetic exchange is facilitated for extrachromosomal DNA, e.g. plasmid-borne antibiotic resistance genes. Consequently, accurate identification of plasmids from whole-genome sequencing (WGS) data remains one of the major challenges for sequencing-based precision medicine in infectious diseases. Here, we assess the heterogeneity of four state-of-the-art tools (cBar, PlasmidFinder, plasmidSPAdes and Recycler) for the in silico prediction of plasmid-derived sequences from WGS data. Heterogeneity, sensitivity and precision were evaluated by reference-independent and reference-dependent benchmarking using 846 Gram-negative clinical isolates. Interestingly, the majority of predicted sequences were tool-specific, resulting in a pronounced heterogeneity across tools for the reference-independent assessment. In the reference-dependent assessment, sensitivity and precision values were found to substantially vary between tools and across taxa, with cBar exhibiting the highest median sensitivity (87.45%) but a low median precision (27.05%). Furthermore, integrating the individual tools into an ensemble approach showed increased sensitivity (95.55%) while reducing the precision (25.62%). CBar and plasmidSPAdes exhibited the strongest concordance with respect to identified antibiotic resistance factors. Moreover, false-positive plasmid predictions typically contained only few antibiotic resistance factors. In conclusion, while high degrees of heterogeneity and variation in sensitivity and precision were observed across the different tools and taxa, existing tools are valuable for investigating the plasmid-borne resistome. Nevertheless, additional studies on representative clinical data sets will be necessary to translate in silico plasmid prediction approaches from research to clinical application.

MOTIVATION: Breast cancer is the second leading cause of cancer death among women. Tumors, even of the same histopathological subtype, exhibit a high genotypic diversity that impedes therapy stratification and that hence must be accounted for in the treatment decision-making process. RESULTS: Here, we present ClinOmicsTrailbc, a comprehensive visual analytics tool for breast cancer decision support that provides a holistic assessment of standard-of-care targeted drugs, candidates for drug repositioning and immunotherapeutic approaches. To this end, our tool analyzes and visualizes clinical markers and (epi-)genomics and transcriptomics datasets to identify and evaluate the tumor's main driver mutations, the tumor mutational burden, activity patterns of core cancer-relevant pathways, drug-specific biomarkers, the status of molecular drug targets and pharmacogenomic influences. In order to demonstrate ClinOmicsTrailbc's rich functionality, we present three case studies highlighting various ways in which ClinOmicsTrailbc can support breast cancer precision medicine. ClinOmicsTrailbc is a powerful integrated visual analytics tool for breast cancer research in general and for therapy stratification in particular, assisting oncologists to find the best possible treatment options for their breast cancer patients based on actionable, evidence-based results. AVAILABILITY AND IMPLEMENTATION: ClinOmicsTrailbc can be freely accessed at https://clinomicstrail.bioinf.uni-sb.de. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

BACKGROUND: In many research disciplines, ordered lists are compared. One example is to compare a subset of all significant genes or proteins in a primary study to those in a replication study. Often, the top of the lists are compared using Venn diagrams, ore more precisely Euler diagrams (set diagrams showing logical relations between a finite collection of different sets). If different cohort sizes, different techniques or algorithms for evaluation were applied, a direct comparison of significant genes with a fixed threshold can however be misleading and approaches comparing lists would be more appropriate. RESULTS: We developed DynaVenn, a web-based tool that incrementally creates all possible subsets from two or three ordered lists and computes for each combination a p-value for the overlap. Respectively, dynamic Venn diagrams are generated as graphical representations. Additionally an animation is generated showing how the most significant overlap is reached by backtracking. We demonstrate the improved performance of DynaVenn over an arbitrary cut-off approach on an Alzheimer's Disease biomarker set. CONCLUSION: DynaVenn combines the calculation of the most significant overlap of different cohorts with an intuitive visualization of the results. It is freely available as a web service at http://www.ccb.uni-saarland.de/dynavenn.

Whole-genome sequencing (WGS) is gaining importance in the analysis of bacterial cultures derived from patients with infectious diseases. Existing computational tools for WGS-based identification have, however, been evaluated on previously defined data relying thereby unwarily on the available taxonomic information.Here, we newly sequenced 846 clinical gram-negative bacterial isolates representing multiple distinct genera and compared the performance of five tools (CLARK, Kaiju, Kraken, DIAMOND/MEGAN and TUIT). To establish a faithful 'gold standard', the expert-driven taxonomy was compared with identifications based on matrix-assisted laser desorption/ionization time-of-flight (MALDI-TOF) mass spectrometry (MS) analysis. Additionally, the tools were also evaluated using a data set of 200 Staphylococcus aureus isolates.CLARK and Kraken (with k =31) performed best with 626 (100%) and 193 (99.5%) correct species classifications for the gram-negative and S. aureus isolates, respectively. Moreover, CLARK and Kraken demonstrated highest mean F-measure values (85.5/87.9% and 94.4/94.7% for the two data sets, respectively) in comparison with DIAMOND/MEGAN (71 and 85.3%), Kaiju (41.8 and 18.9%) and TUIT (34.5 and 86.5%). Finally, CLARK, Kaiju and Kraken outperformed the other tools by a factor of 30 to 170 fold in terms of runtime.We conclude that the application of nucleotide-based tools using k-mers-e.g. CLARK or Kraken-allows for accurate and fast taxonomic characterization of bacterial isolates from WGS data. Hence, our results suggest WGS-based genotyping to be a promising alternative to the MS-based biotyping in clinical settings. Moreover, we suggest that complementary information should be used for the evaluation of taxonomic classification tools, as public databases may suffer from suboptimal annotations.

Motivation: Although the amount of small non-coding RNA-sequencing data is continuously increasing, it is still unclear to which extent small RNAs are represented in the human genome. Results: In this study we analyzed 303 billion sequencing reads from nearly 25 000 datasets to answer this question. We determined that 0.8% of the human genome are reliably covered by 874 123 regions with an average length of 31 nt. On the basis of these regions, we found that among the known small non-coding RNA classes, microRNAs were the most prevalent. In subsequent steps, we characterized variations of miRNAs and performed a staged validation of 11 877 candidate miRNAs. Of these, many were actually expressed and significantly dysregulated in lung cancer. Selected candidates were finally validated by northern blots. Although isolated miRNAs could still be present in the human genome, our presented set likely contains the largest fraction of human miRNAs. Contact: [email protected] or [email protected]. Supplementary information: Supplementary data are available at Bioinformatics online.

Motivation: Transcriptional regulators play a major role in most biological processes. Alterations in their activities are associated with a variety of diseases and in particular with tumor development and progression. Hence, it is important to assess the effects of deregulated regulators on pathological processes. Results: Here, we present REGulator-Gene Association Enrichment (REGGAE), a novel method for the identification of key transcriptional regulators that have a significant effect on the expression of a given set of genes, e.g. genes that are differentially expressed between two sample groups. REGGAE uses a Kolmogorov-Smirnov-like test statistic that implicitly combines associations between regulators and their target genes with an enrichment approach to prioritize the influence of transcriptional regulators. We evaluated our method in two different application scenarios, which demonstrate that REGGAE is well suited for uncovering the influence of transcriptional regulators and is a valuable tool for the elucidation of complex regulatory mechanisms. Availability and implementation: REGGAE is freely available at https://regulatortrail.bioinf.uni-sb.de. Supplementary information: Supplementary data are available at Bioinformatics online.

UNLABELLED: In medical research, it is crucial to understand the functional consequences of genetic alterations, for example, non-synonymous single nucleotide variants (nsSNVs). NsSNVs are known to be causative for several human diseases. However, the genetic basis of complex disorders such as diabetes or cancer comprises multiple factors. Methods to analyze putative synergetic effects of multiple such factors, however, are limited. Here, we concentrate on nsSNVs and present BALL-SNPgp, a tool for structural and functional characterization of nsSNVs, which is aimed to improve pathogenicity assessment in computational diagnostics. Based on annotated SNV data, BALL-SNPgp creates a three-dimensional visualization of the encoded protein, collects available information from different resources concerning disease relevance and other functional annotations, performs cluster analysis, predicts putative binding pockets and provides data on known interaction sites. AVAILABILITY AND IMPLEMENTATION: BALL-SNPgp is based on the comprehensive C ++ framework Biochemical Algorithms Library (BALL) and its visualization front-end BALLView. Our tool is available at www.ccb.uni-saarland.de/BALL-SNPgp CONTACT: [email protected].

MOTIVATION: Gene set analysis has revolutionized the interpretation of high-throughput transcriptomic data. Nowadays, with comprehensive studies that measure multiple -omics from the same sample, powerful tools for the integrative analysis of multi-omics datasets are required. RESULTS: Here, we present GeneTrail2, a web service allowing the integrated analysis of transcriptomic, miRNomic, genomic and proteomic datasets. It offers multiple statistical tests, a large number of predefined reference sets, as well as a comprehensive collection of biological categories and enables direct comparisons between the computed results. We used GeneTrail2 to explore pathogenic mechanisms of Wilms tumors. We not only succeeded in revealing signaling cascades that may contribute to the malignancy of blastemal subtype tumors but also identified potential biomarkers for nephroblastoma with adverse prognosis. The presented use-case demonstrates that GeneTrail2 is well equipped for the integrative analysis of comprehensive -omics data and may help to shed light on complex pathogenic mechanisms in cancer and other diseases. AVAILABILITY AND IMPLEMENTATION: GeneTrail2 can be freely accessed under https://genetrail2.bioinf.uni-sb.de CONTACT: : [email protected] SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

Non-synonymous single nucleotide variants (nsSNVs) in coding DNA regions can result in phenotypic differences between individuals; however, only some nsSNVs are causative for a certain disease. As just a fraction of respective nsSNVs is annotated in databases, computational biology tools are applied to predict the pathogenicity in silico. In addition to applications in oncology, novel molecular diagnostic tests have been developed for cardiovascular disorders as a leading cause of morbidity and mortality in industrialized nations. We explored the concordance and performance of 13 nsSNV pathogenicity prediction tools on panel sequencing results of dilated cardiomyopathy. The analyzed data set from the INHERITANCE study contained 842 nsSNVs discovered in 639 patients, screened for the full sequence of 76 genes related to cardiomyopathies. The single tools prediction revealed a surprisingly high heterogeneity and discordance based on the implemented prediction method. Known disease associations were not reported by the tools, limiting usability in clinics. Because different tools have different advantages, we combined their results. By clustering of correlated methods using similar prediction strategies and calculating a majority vote-based consensus, we found that the prediction accuracy and sensitivity can be further improved. Although challenges remain, different in silico tools bear the potential to predict the malignancy of nsSNVs, especially if different algorithms are combined. Most tools rely mainly on sequence features; beyond these, structural information is important to analyze the relationship of nsSNVs with disease phenotypes. Likewise, current tools consider single nsSNVs, which may, however, show a cumulative effect and turn neutral mutations in an ensemble into pathogenic variants.

UNLABELLED: The deregulation of biochemical pathways plays a central role in many diseases like cancer or Parkinsons's disease. In silico tools for calculating these deregulated pathways may help to gain new insights into pathogenic mechanisms and may open novel avenues for therapy stratification in the sense of personalized medicine. Here, we present NetworkTrail, a web service for the detection of deregulated pathways and subgraphs in biological networks. NetworkTrail uses a state-of-the-art integer linear programming-based approach for this task and offers interfaces to the Biological Network Analyzer (BiNA) and Cytoscape Web for visualizing the resulting subnetworks. By providing an accessible interface to otherwise hard-to-use command line tools, the new web service enables non-experts to quickly and reliably carry out this type of network analyses. AVAILABILITY AND IMPLEMENTATION: NetworkTrail is a JavaServer Pages-based web service. The algorithm for finding deregulated subnetworks has been implemented in C++. NetworkTrail is available at http://networktrail.bioinf.uni-sb.de/.

BACKGROUND: Expression profiling provides new insights into regulatory and metabolic processes and in particular into pathogenic mechanisms associated with diseases. Besides genes, non-coding transcripts as microRNAs (miRNAs) gained increasing relevance in the last decade. To understand the regulatory processes of miRNAs on genes, integrative computer-aided approaches are essential, especially in the light of complex human diseases as cancer. RESULTS: Here, we present miRTrail, an integrative tool that allows for performing comprehensive analyses of interactions of genes and miRNAs based on expression profiles. The integrated analysis of mRNA and miRNA data should generate more robust and reliable results on deregulated pathogenic processes and may also offer novel insights into the regulatory interactions between miRNAs and genes. Our web-server excels in carrying out gene sets analysis, analysis of miRNA sets as well as the combination of both in a systems biology approach. To this end, miRTrail integrates information on 20.000 genes, almost 1.000 miRNAs, and roughly 280.000 putative interactions, for Homo sapiens and accordingly for Mus musculus and Danio rerio. The well-established, classical Chi-squared test is one of the central techniques of our tool for the joint consideration of miRNAs and their targets. For interactively visualizing obtained results, it relies on the network analyzers and viewers BiNA or Cytoscape-web, also enabling direct access to relevant literature. We demonstrated the potential of miRTrail by applying our tool to mRNA and miRNA data of malignant melanoma. MiRTrail identified several deregulated miRNAs that target deregulated mRNAs including miRNAs hsa-miR-23b and hsa-miR-223, which target the highest numbers of deregulated mRNAs and regulate the pathway "basal cell carcinoma". In addition, both miRNAs target genes like PTCH1 and RASA1 that are involved in many oncogenic processes. CONCLUSIONS: The application on melanoma samples demonstrates that the miRTrail platform may open avenues for investigating the regulatory interactions between genes and miRNAs for a wide range of human diseases. Moreover, miRTrail cannot only be applied to microarray based expression profiles, but also to NGS-based transcriptomic data. The program is freely available as web-server at mirtrail.bioinf.uni-sb.de.

Handwritten word spotting aims at making document images amenable to browsing and searching by keyword retrieval. In this paper, we present a word spotting system based on Hidden Markov Models (HMM) that uses trained subword models to spot keywords. With the proposed method, arbitrary keywords can be spotted that do not need to be present in the training set. Also, no text line segmentation is required. On the modern IAM off-line database and the historical George Washington database we show that the proposed system outperforms a standard template matching approach based on dynamic time warping (DTW).

MOTIVATION: Deregulated signaling cascades are known to play a crucial role in many pathogenic processes, among them are tumor initiation and progression. In the recent past, modern experimental techniques that allow for measuring the amount of mRNA transcripts of almost all known human genes in a tissue or even in a single cell have opened new avenues for studying the activity of the signaling cascades and for understanding the information flow in the networks. RESULTS: We present a novel dynamic programming algorithm for detecting deregulated signaling cascades. The so-called FiDePa (Finding Deregulated Paths) algorithm interprets differences in the expression profiles of tumor and normal tissues. It relies on the well-known gene set enrichment analysis (GSEA) and efficiently detects all paths in a given regulatory or signaling network that are significantly enriched with differentially expressed genes or proteins. Since our algorithm allows for comparing a single tumor expression profile with the control group, it facilitates the detection of specific regulatory features of a tumor that may help to optimize tumor therapy. To demonstrate the capabilities of our algorithm, we analyzed a glioma expression dataset with respect to a directed graph that combined the regulatory networks of the KEGG and TRANSPATH database. The resulting glioma consensus network that encompasses all detected deregulated paths contained many genes and pathways that are known to be key players in glioma or cancer-related pathogenic processes. Moreover, we were able to correlate clinically relevant features like necrosis or metastasis with the detected paths. AVAILABILITY: C++ source code is freely available, BiNA can be downloaded from http://www.bnplusplus.org/. CONTACT: [email protected] SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

BACKGROUND: High-throughput methods that allow for measuring the expression of thousands of genes or proteins simultaneously have opened new avenues for studying biochemical processes. While the noisiness of the data necessitates an extensive pre-processing of the raw data, the high dimensionality requires effective statistical analysis methods that facilitate the identification of crucial biological features and relations. For these reasons, the evaluation and interpretation of expression data is a complex, labor-intensive multi-step process. While a variety of tools for normalizing, analysing, or visualizing expression profiles has been developed in the last years, most of these tools offer only functionality for accomplishing certain steps of the evaluation pipeline. RESULTS: Here, we present a web-based toolbox that provides rich functionality for all steps of the evaluation pipeline. Our tool GeneTrailExpress offers besides standard normalization procedures powerful statistical analysis methods for studying a large variety of biological categories and pathways. Furthermore, an integrated graph visualization tool, BiNA, enables the user to draw the relevant biological pathways applying cutting-edge graph-layout algorithms. CONCLUSION: Our gene expression toolbox with its interactive visualization of the pathways and the expression values projected onto the nodes will simplify the analysis and interpretation of biochemical pathways considerably.

BACKGROUND: Gene Set Enrichment Analysis (GSEA) is a computational method for the statistical evaluation of sorted lists of genes or proteins. Originally GSEA was developed for interpreting microarray gene expression data, but it can be applied to any sorted list of genes. Given the gene list and an arbitrary biological category, GSEA evaluates whether the genes of the considered category are randomly distributed or accumulated on top or bottom of the list. Usually, significance scores (p-values) of GSEA are computed by nonparametric permutation tests, a time consuming procedure that yields only estimates of the p-values. RESULTS: We present a novel dynamic programming algorithm for calculating exact significance values of unweighted Gene Set Enrichment Analyses. Our algorithm avoids typical problems of nonparametric permutation tests, as varying findings in different runs caused by the random sampling procedure. Another advantage of the presented dynamic programming algorithm is its runtime and memory efficiency. To test our algorithm, we applied it not only to simulated data sets, but additionally evaluated expression profiles of squamous cell lung cancer tissue and autologous unaffected tissue.

BACKGROUND: The development of effective frameworks that permit an accurate diagnosis of tumors, especially in their early stages, remains a grand challenge in the field of bioinformatics. Our approach uses statistical learning techniques applied to multiple antigen tumor antigen markers utilizing the immune system as a very sensitive marker of molecular pathological processes. For validation purposes we choose the intracranial meningioma tumors as model system since they occur very frequently, are mostly benign, and are genetically stable. RESULTS: A total of 183 blood samples from 93 meningioma patients (WHO stages I-III) and 90 healthy controls were screened for seroreactivity with a set of 57 meningioma-associated antigens. We tested several established statistical learning methods on the resulting reactivity patterns using 10-fold cross validation. The best performance was achieved by Naïve Bayes Classifiers. With this classification method, our framework, called Minimally Invasive Multiple Marker (MIMM) approach, yielded a specificity of 96.2%, a sensitivity of 84.5%, and an accuracy of 90.3%, the respective area under the ROC curve was 0.957. Detailed analysis revealed that prediction performs particularly well on low-grade (WHO I) tumors, consistent with our goal of early stage tumor detection. For these tumors the best classification result with a specificity of 97.5%, a sensitivity of 91.3%, an accuracy of 95.6%, and an area under the ROC curve of 0.971 was achieved using a set of 12 antigen markers only. This antigen set was detected by a subset selection method based on Mutual Information. Remarkably, our study proves that the inclusion of non-specific antigens, detected not only in tumor but also in normal sera, increases the performance significantly, since non-specific antigens contribute additional diagnostic information. CONCLUSION: Our approach offers the possibility to screen members of risk groups as a matter of routine such that tumors hopefully can be diagnosed immediately after their genesis. The early detection will finally result in a higher cure- and lower morbidity-rate.