Software defect prediction using learning to rank approach.

Software defect prediction (SDP) plays a significant role in detecting the most likely defective software modules and optimizing the allocation of testing resources. In practice, though, project managers must not only identify defective modules, but also rank them in a specific order to optimize the resource allocation and minimize testing costs, especially for projects with limited budgets. This vital task can be accomplished using Learning to Rank (LTR) algorithm. This algorithm is a type of machine learning methodology that pursues two important tasks: prediction and learning. Although this algorithm is commonly used in information retrieval, it also presents high efficiency for other problems, like SDP. The LTR approach is mainly used in defect prediction to predict and rank the most likely buggy modules based on their bug count or bug density. This research paper conducts a comprehensive comparison study on the behavior of eight selected LTR models using two target variables: bug count and bug density. It also studies the effect of using imbalance learning and feature selection on the employed LTR models. The models are empirically evaluated using Fault Percentile Average. Our results show that using bug count as ranking criteria produces higher scores and more stable results across multiple experiment settings. Moreover, using imbalance learning has a positive impact for bug density, but on the other hand it leads to a negative impact for bug count. Lastly, using the feature selection does not show significant improvement for bug density, while there is no impact when bug count is used. Therefore, we conclude that using feature selection and imbalance learning with LTR does not come up with superior or significant results.

A personalized stepwise dynamic predictive algorithm of the time to first treatment in chronic lymphocytic leukemia.

Personalized prediction is ideal in chronic lymphocytic leukemia (CLL). Although refined models have been developed, stratifying patients in risk groups, it is required to accommodate time-dependent information of patients, to address the clinical heterogeneity observed within these groups. In this direction, this study proposes a personalized stepwise dynamic predictive algorithm (PSDPA) for the time-to-first-treatment of the individual patient. The PSDPA introduces a personalized Score, reflecting the evolution in the patient's follow-up, employed to develop a reference pool of patients. Score evolution's similarity is used to predict, at a selected time point, the time-to-first-treatment for a new patient. Additional patient's biological information may be utilized. The algorithm was applied to 20 CLL patients, indicating that stricter assessment criteria for the Score evolution's similarity, and biological similarity exploitation, may improve prediction. The PSDPA capitalizes on both the follow-up and the biological background of the individual patient, dynamically promoting personalized prediction in CLL.

Single Nucleotide Polymorphisms' Causal Structure Robustness within Coronary Artery Disease Patients.

An ever-growing amount of accumulated data has materialized in several scientific fields, due to recent technological progress. New challenges emerge in exploiting these data and utilizing the valuable available information. Causal models are a powerful tool that can be employed towards this aim, by unveiling the structure of causal relationships between different variables. The causal structure may avail experts to better understand relationships, or even uncover new knowledge. Based on 963 patients with coronary artery disease, the robustness of the causal structure of single nucleotide polymorphisms was assessed, taking into account the value of the Syntax Score, an index that evaluates the complexity of the disease. The causal structure was investigated, both locally and globally, under different levels of intervention, reflected in the number of patients that were randomly excluded from the original datasets corresponding to two categories of the Syntax Score, zero and positive. It is shown that the causal structure of single nucleotide polymorphisms was more robust under milder interventions, whereas in the case of stronger interventions, the impact increased. The local causal structure around the Syntax Score was studied in the case of a positive Syntax Score, and it was found to be resilient, even when the intervention was strong. Consequently, employing causal models in this context may increase the understanding of the biological aspects of coronary artery disease.

A wide foodomics approach coupled with metagenomics elucidates the environmental signature of potatoes.

The term "terroir" has been widely employed to link differential geographic phenotypes with sensorial signatures of agricultural food products, influenced by agricultural practices, soil type, and climate. Nowadays, the geographical indications labeling has been developed to safeguard the quality of plant-derived food that is linked to a certain terroir and is generally considered as an indication of superior organoleptic properties. As the dynamics of agroecosystems are highly intricate, consisting of tangled networks of interactions between plants, microorganisms, and the surrounding environment, the recognition of the key molecular components of terroir fingerprinting remains a great challenge to protect both the origin and the safety of food commodities. Furthermore, the contribution of microbiome as a potential driver of the terroir signature has been underestimated. Herein, we present a first comprehensive view of the multi-omic landscape related to transcriptome, proteome, epigenome, and metagenome of the popular Protected Geographical Indication potatoes of Naxos.

Disclosing the molecular basis of salinity priming in olive trees using proteogenomic model discovery.

Plant responses to salinity are becoming increasingly understood, however, salt priming mechanisms remain unclear, especially in perennial fruit trees. Herein, we showed that low-salt pre-exposure primes olive (Olea europaea) plants against high salinity stress. We then performed a proteogenomic study to characterize priming responses in olive roots and leaves. Integration of transcriptomic and proteomic data along with metabolic data revealed robust salinity changes that exhibit distinct or overlapping patterns in olive tissues, among which we focused on sugar regulation. Using the multi-crossed -omics data set, we showed that major differences between primed and nonprimed tissues are mainly associated with hormone signaling and defense-related interactions. We identified multiple genes and proteins, including known and putative regulators, that reported significant proteomic and transcriptomic changes between primed and nonprimed plants. Evidence also supported the notion that protein post-translational modifications, notably phosphorylations, carbonylations and S-nitrosylations, promote salt priming. The proteome and transcriptome abundance atlas uncovered alterations between mRNA and protein quantities within tissues and salinity conditions. Proteogenomic-driven causal model discovery also unveiled key interaction networks involved in salt priming. Data generated in this study are important resources for understanding salt priming in olive tree and facilitating proteogenomic research in plant physiology.

The Implication of X-Linked Genetic Polymorphisms in Susceptibility and Sexual Dimorphism of Cancer.

The X-chromosome is implicated in cancer development through various mechanisms, including X-inactivation defects, loss of heterozygosity, and germline and somatic alterations of X-linked genes. Sex is a key factor which influences cancer susceptibility as many cancer types show sexual dimorphism in their incidence. The aim of this review was to summarize the germline genetic polymorphisms lying on the X-chromosome that have been associated with cancer susceptibility and to evaluate their possible implication in cancer-related sexual dimorphism. PubMed and Web of Science were searched using the terms "X-chromosome", "polymorphism" and "cancer". The literature review revealed 39 articles reporting 33 genetic variants in 22 X-linked genes as being associated with cancer risk. Most of these genes interact with each other in a direct or indirect way, as GeneMANIA software revealed, demonstrating the complication of the mechanisms through which they are involved in tumorigenesis. Polymorphisms in eight genes [androgen receptor (AR), fibroblast growth factor 13 (FGF13), forkhead box P3 (FOXP3), L1 cell adhesion molecule (L1CAM), nudix hydrolase 11 (NUDT11), Shroom family member 2 (SHROOM2), transcription elongation factor A-like 7 (TCEAL7) and TIMP metallopeptidase inhibitor 1 (TIMP1)] are reported to have a sex-specific association with cancer susceptibility, which might explain the sexual dimorphism of certain cancer types. All of the above eight mentioned genes, with the exception of L1CAM, exhibit differences in their expression pattern between breast tumor (sex-specific)/thyroid tumor (sex-influenced) vs. normal tissues according to our analysis using GENT2 software. Additionally, differences in breast or thyroid tumor compared with normal tissues were also observed in five genes analyzed with GENT2 software that were previously related to sex-influenced cancer according to literature. Finally, the present review points out the need for the development of appropriate free and user-friendly statistical software in order to reduce bias/errors in statistical analyses and overcome researchers' reluctance to include X-chromosome variants in their genetic-association studies.

Dissecting miRNA-Gene Networks to Map Clinical Utility Roads of Pharmacogenomics-Guided Therapeutic Decisions in Cardiovascular Precision Medicine.

MicroRNAs (miRNAs) create systems networks and gene-expression circuits through molecular signaling and cell interactions that contribute to health imbalance and the emergence of cardiovascular disorders (CVDs). Because the clinical phenotypes of CVD patients present a diversity in their pathophysiology and heterogeneity at the molecular level, it is essential to establish genomic signatures to delineate multifactorial correlations, and to unveil the variability seen in therapeutic intervention outcomes. The clinically validated miRNA biomarkers, along with the relevant SNPs identified, have to be suitably implemented in the clinical setting in order to enhance patient stratification capacity, to contribute to a better understanding of the underlying pathophysiological mechanisms, to guide the selection of innovative therapeutic schemes, and to identify innovative drugs and delivery systems. In this article, the miRNA-gene networks and the genomic signatures resulting from the SNPs will be analyzed as a method of highlighting specific gene-signaling circuits as sources of molecular knowledge which is relevant to CVDs. In concordance with this concept, and as a case study, the design of the clinical trial GESS (NCT03150680) is referenced. The latter is presented in a manner to provide a direction for the improvement of the implementation of pharmacogenomics and precision cardiovascular medicine trials.

An empirical study of COVID-19 related posts on Stack Overflow: Topics and technologies.

The COVID-19 outbreak, also known as the coronavirus pandemic, has left its mark on every aspect of our lives and at the time of this writing is still an ongoing battle. Beyond the immediate global-wide health response, the pandemic has triggered a significant number of IT initiatives to track, visualize, analyze and potentially mitigate the phenomenon. For individuals or organizations interested in developing COVID-19 related software, knowledge-sharing communities such as Stack Overflow proved to be an effective source of information for tackling commonly encountered problems. As an additional contribution to the investigation of this unprecedented health crisis and to assess how fast and how well the community of developers has responded, we performed a study on COVID-19 related posts in Stack Overflow. In particular, we profiled relevant questions based on key post features and their evolution, identified the most prominent technologies adopted for developing COVID-19 software and their interrelations and focused on the most persevering problems faced by developers. For the analysis of posts we employed descriptive statistics, Association Rule Graphs, Survival Analysis and Latent Dirichlet Allocation. The results reveal that the response of the developers' community to the pandemic was immediate and that the interest of developers on COVID-19 related challenges was sustained after its initial peak. In terms of the problems addressed, the results show a clear focus on COVID-19 data collection, analysis and visualization from/to the web, in line with the general needs for monitoring the pandemic.

A Strong Seasonality Pattern for Covid-19 Incidence Rates Modulated by UV Radiation Levels.

The Covid-19 pandemic has required nonpharmaceutical interventions, primarily physical distancing, personal hygiene and face mask use, to limit community transmission, irrespective of seasons. In fact, the seasonality attributes of this pandemic remain one of its biggest unknowns. Early studies based on past experience from respiratory diseases focused on temperature or humidity, with disappointing results. Our hypothesis that ultraviolet (UV) radiation levels might be a factor and a more appropriate parameter has emerged as an alternative to assess seasonality and exploit it for public health policies. Using geographical, socioeconomic and epidemiological criteria, we selected twelve North-equatorial-South countries with similar characteristics. We then obtained UV levels, mobility and Covid-19 daily incidence rates for nearly the entire 2020. Using machine learning, we demonstrated that UV radiation strongly associated with incidence rates, more so than mobility did, indicating that UV is a key seasonality indicator for Covid-19, irrespective of the initial conditions of the epidemic. Our findings can inform the implementation of public health emergency measures, partly based on seasons in the Northern and Southern Hemispheres, as the pandemic unfolds into 2021.

A Risk-Stratification Machine Learning Framework for the Prediction of Coronary Artery Disease Severity: Insights From the GESS Trial.

Our study aims to develop a data-driven framework utilizing heterogenous electronic medical and clinical records and advanced Machine Learning (ML) approaches for: (i) the identification of critical risk factors affecting the complexity of Coronary Artery Disease (CAD), as assessed via the SYNTAX score; and (ii) the development of ML prediction models for accurate estimation of the expected SYNTAX score. We propose a two-part modeling technique separating the process into two distinct phases: (a) a binary classification task for predicting, whether a patient is more likely to present with a non-zero SYNTAX score; and (b) a regression task to predict the expected SYNTAX score accountable to individual patients with a non-zero SYNTAX score. The framework is based on data collected from the GESS trial (NCT03150680) comprising electronic medical and clinical records for 303 adult patients with suspected CAD, having undergone invasive coronary angiography in AHEPA University Hospital of Thessaloniki, Greece. The deployment of the proposed approach demonstrated that atherogenic index of plasma levels, diabetes mellitus and hypertension can be considered as important risk factors for discriminating patients into zero- and non-zero SYNTAX score groups, whereas diastolic and systolic arterial blood pressure, peripheral vascular disease and body mass index can be considered as significant risk factors for providing an accurate estimation of the expected SYNTAX score, given that a patient belongs to the non-zero SYNTAX score group. The experimental findings utilizing the identified set of important risk factors indicate a sufficient prediction performance for the Support Vector Machine model (classification task) with an F-measure score of ~0.71 and the Support Vector Regression model (regression task) with a median absolute error value of ~6.5. The proposed data-driven framework described herein present evidence of the prediction capacity and the potential clinical usefulness of the developed risk-stratification models. However, further experimentation in a larger clinical setting is needed to ensure the practical utility of the presented models in a way to contribute to a more personalized management and counseling of CAD patients.

Could Causal Discovery in Proteogenomics Assist in Understanding Gene-Protein Relations? A Perennial Fruit Tree Case Study Using Sweet Cherry as a Model.

Genome-wide transcriptome analysis is a method that produces important data on plant biology at a systemic level. The lack of understanding of the relationships between proteins and genes in plants necessitates a further thorough analysis at the proteogenomic level. Recently, our group generated a quantitative proteogenomic atlas of 15 sweet cherry (Prunus avium L.) cv. 'Tragana Edessis' tissues represented by 29,247 genes and 7584 proteins. The aim of the current study was to perform a targeted analysis at the gene/protein level to assess the structure of their relation, and the biological implications. Weighted correlation network analysis and causal modeling were employed to, respectively, cluster the gene/protein pairs, and reveal their cause-effect relations, aiming to assess the associated biological functions. To the best of our knowledge, this is the first time that causal modeling has been employed within the proteogenomics concept in plants. The analysis revealed the complex nature of causal relations among genes/proteins that are important for traits of interest in perennial fruit trees, particularly regarding the fruit softening and ripening process in sweet cherry. Causal discovery could be used to highlight persistent relations at the gene/protein level, stimulating biological interpretation and facilitating further study of the proteogenomic atlas in plants.

The Coming of Age for Big Data in Systems Radiobiology, an Engineering Perspective.

As high-throughput approaches in biological and biomedical research are transforming the life sciences into information-driven disciplines, modern analytics platforms for big data have started to address the needs for efficient and systematic data analysis and interpretation. We observe that radiobiology is following this general trend, with -omics information providing unparalleled depth into the biomolecular mechanisms of radiation response-defined as systems radiobiology. We outline the design of computational frameworks and discuss the analysis of big data in low-dose ionizing radiation (LDIR) responses of the mammalian brain. Following successful examples and best practices of approaches for the analysis of big data in life sciences and health care, we present the needs and requirements for radiation research. Our goal is to raise awareness for the radiobiology community about the new technological possibilities that can capture complex information and execute data analytics on a large scale. The production of large data sets from genome-wide experiments (quantity) and the complexity of radiation research with multidimensional experimental designs (quality) will necessitate the adoption of latest information technologies. The main objective was to translate research results into applied clinical and epidemiological practice and understand the responses of biological tissues to LDIR to define new radiation protection policies. We envisage a future where multidisciplinary teams include data scientists, artificial intelligence experts, DevOps engineers, and of course radiation experts to fulfill the augmented needs of the radiobiology community, accelerate research, and devise new strategies.

Sequence variation, common tissue expression patterns and learning models: a genome-wide survey of vertebrate ribosomal proteins.

Ribosomal genes produce the constituents of the ribosome, one of the most conserved subcellular structures of all cells, from bacteria to eukaryotes, including animals. There are notions that some protein-coding ribosomal genes vary in their roles across species, particularly vertebrates, through the involvement of some in a number of genetic diseases. Based on extensive sequence comparisons and systematic curation, we establish a reference set for ribosomal proteins (RPs) in eleven vertebrate species and quantify their sequence conservation levels. Moreover, we correlate their coordinated gene expression patterns within up to 33 tissues and assess the exceptional role of paralogs in tissue specificity. Importantly, our analysis supported by the development and use of machine learning models strongly proposes that the variation in the observed tissue-specific gene expression of RPs is rather species-related and not due to tissue-based evolutionary processes. The data obtained suggest that RPs exhibit a complex relationship between their structure and function that broadly maintains a consistent expression landscape across tissues, while most of the variation arises from species idiosyncrasies. The latter may be due to evolutionary change and adaptation, rather than functional constraints at the tissue level throughout the vertebrate lineage.

Investigation of the relationship between sleep disorders and xerostomia.

OBJECTIVES: To investigate the relationship between sleep disorders, morning hyposalivation, and subjective feeling of dry mouth. MATERIALS AND METHODS: A cross-sectional, observational, clinical study was carried out in a homogenous population sample which consists of Greek male soldiers without any medical history. After the application of oral modified Schirmer test, the sample was divided into a study group (n = 63) (MST < 25 mm/3 min) and a control group (n = 110) (MST ≥ 25 mm/3 min). In order to assess daytime sleepiness, risk of obstructive sleep apnea (OSA), sleep quality, sleep bruxism (SB), and subjective feeling of dry mouth, all the participants filled in the following scales in Greek version: Epworth Sleepiness Scale (ESS), Pittsburgh Sleep Quality Index (PSQI), Berlin Questionnaire (BQ), a SB questionnaire, and Xerostomia Inventory (XI) respectively. In every subgroup that came of ESS, PSQI, BQ, and SB questionnaire scoring, subjective feeling of dry mouth was evaluated, based on XI values. RESULTS: Statistically significant difference (p < 0.001) through PSQI scores was found between the study and control group. In contrast, a statistically significant difference was not obtained for the scores of ESS (p = 0.293), BQ (p = 0.089), and SB questionnaire (p = 0.730). XI scores introduced statistically significant difference between the subgroups of PSQI (p < 0.001), BQ (p = 0.001), SB questionnaire (p = 0.004)  and statistically weak between the subgroups of ESS (p = 0.049). CONCLUSIONS: This is the first research study so far suggesting that patients with morning hyposalivation exhibit poor sleep quality using an objective method. The present results have, also, shown that subjective feeling of dry mouth is related to excessive daytime sleepiness, poor sleep quality, high risk of obstructive sleep apnea, and sleep bruxism, but larger-scale studies are still needed. CLINICAL RELEVANCE: These findings should keep dentists aware of a possible association between xerostomia and sleep disorders and support larger-scale studies.

Study of gene expressions' correlation structures in subgroups of Chronic Lymphocytic Leukemia Patients.

In chronic lymphocytic leukemia (CLL) the interaction of leukemic cells with the microenvironment ultimately affects patient outcome. CLL cases can be divided in two subgroups with different clinical course based on the mutational status of the immunoglobulin heavy variable (IGHV) genes: mutated CLL (M-CLL) and unmutated CLL (U-CLL). Since in CLL, the differentiated relation of genes between the two subgroups is of greater importance than the individual gene behavior, this paper investigates the differences between the groups' gene interactions, by comparing their correlation structures. Fisher's test and Zou's confidence intervals are employed to detect differences of correlation coefficients. Afterwards, networks created by the genes participating in most differences are estimated with the use of structural equation models (SEM). The analysis is enhanced with graph modeling in order to visualize the between group differences in the gene structures of the two subgroups. The applied methodology revealed stronger correlations between genes in U-CLL patients, a finding in line with related biomedical literature. Using SEM for multigroup analysis, different gene structures between the two groups of patients were confirmed. The study of correlation structures can facilitate the detection of differential gene expression profiles in CLL subgroups, with potential applications in other diseases. Comparison of correlations can be very useful in understanding the complex internal structural differences which signify the variations of a disease.

Temporal Trends in Chronic Total Occlusion Interventions in Europe.

BACKGROUND: The study focuses on the evolution of practice, procedural outcomes, and in-hospital complications of chronic total occlusion percutaneous coronary intervention in Europe. METHODS AND RESULTS: Data from 17 626 procedures enrolled in European Registry of Chronic Total Occlusion between January 2008 and June 2015 were assessed. The mean patient age was 63.9±10.9 years; 85% were men. Procedural success increased from 79.7% to 89.3% through the study period. Patients enrolled during the years had increasing comorbidities and lesion complexity (J-CTO score [Multicenter CTO Registry of Japan] increased from 1.76±1.03 in 2008 to 2.17±0.91 in 2015; P for trend, <0.001). Retrograde approach utilization steadily increased from 10.1% in 2008 to 29.9% in 2015 ( P for trend, <0.001). Antegrade dissection reentry adoption was low, not exceeding 5.5%. In-hospital mortality decreased during the study period from 0.4% to 0.1% ( P for trend, <0.001), whereas in-hospital complication rates remained essentially unchanged, in the range 4.4% to 5.2% ( P for trend, 0.390). CONCLUSIONS: Chronic total occlusion percutaneous coronary intervention has shown a steady increase in procedural success rate over time, with unchanged complication rates, despite the increasing complexity of the lesions attempted. The J-CTO score predictive value for procedural success was low for the entire registry and had no predictive ability for the retrograde approach.

Human papillomavirus E7 protein detection as a method of triage to colposcopy of HPV positive women, in comparison to genotyping and cytology. Final results of the PIPAVIR study.

The objective of the presented cross-sectional-evaluation-screening study is the clinical evaluation of high-risk(hr)HPVE7-protein detection as a triage method to colposcopy for hrHPV-positive women, using a newly developed sandwich-ELISA-assay. Between 2013-2015, 2424 women, 30-60 years old, were recruited at the Hippokratio Hospital, Thessaloniki/Greece and the Im Mare Klinikum, Kiel/Germany, and provided a cervical sample used for Liquid Based Cytology, HPV DNA genotyping, and E7 detection using five different E7-assays: "recomWell HPV16/18/45KJhigh", "recomWell HPV16/18/45KJlow", "recomWell HPV39/51/56/59", "recomWell HPV16/31/33/35/52/58" and "recomWell HPVHRscreen" (for 16,18,31,33,35,39,45,51,52,56,58,59 E7), corresponding to different combinations of hrHPVE7-proteins. Among 1473 women with eligible samples, those positive for cytology (ASCUS+ 7.2%), and/or hrHPV DNA (19.1%) were referred for colposcopy. Cervical Intraepithelial Neoplasia grade 2 or worse (CIN2+) was detected in 27 women (1.8%). For HPV16/18-positive women with no triage, sensitivity, positive predictive value (PPV) and the number of colposcopies needed to detect one case of CIN2+ were 100.0%, 11.11% and 9.0 respectively. The respective values for E7-testing as a triage method to colposcopy ranged from 75.0-100.0%, 16.86-26.08% and 3.83-5.93. Sensitivity and PPV for cytology as triage for hrHPV(non16/18)-positive women were 45.45% and 27.77%; for E7 test the respective values ranged from 72.72-100.0% and 16.32-25.0%. Triage of HPV 16/18-positive women to colposcopy with the E7 test presents better performance than no triage, decreasing the number of colposcopies needed to detect one CIN2+. In addition, triage of hrHPV(non16/18)-positive women with E7 test presents better sensitivity and slightly worse PPV than cytology, a fact that advocates for a full molecular screening approach.

Not all IGHV3-21 chronic lymphocytic leukemias are equal: prognostic considerations.

An unresolved issue in chronic lymphocytic leukemia (CLL) is whether IGHV3-21 gene usage, in general, or the expression of stereotyped B-cell receptor immunoglobulin defining subset #2 (IGHV3-21/IGLV3-21), in particular, determines outcome for IGHV3-21-utilizing cases. We reappraised this issue in 8593 CLL patients of whom 437 (5%) used the IGHV3-21 gene with 254/437 (58%) classified as subset #2. Within subset #2, immunoglobulin heavy variable (IGHV)-mutated cases predominated, whereas non-subset #2/IGHV3-21 was enriched for IGHV-unmutated cases (P = .002). Subset #2 exhibited significantly shorter time-to-first-treatment (TTFT) compared with non-subset #2/IGHV3-21 (22 vs 60 months, P = .001). No such difference was observed between non-subset #2/IGHV3-21 vs the remaining CLL with similar IGHV mutational status. In conclusion, IGHV3-21 CLL should not be axiomatically considered a homogeneous entity with adverse prognosis, given that only subset #2 emerges as uniformly aggressive, contrasting non-subset #2/IGVH3-21 patients whose prognosis depends on IGHV mutational status as the remaining CLL.

Clinical effect of stereotyped B-cell receptor immunoglobulins in chronic lymphocytic leukaemia: a retrospective multicentre study.

BACKGROUND: About 30% of cases of chronic lymphocytic leukaemia (CLL) carry quasi-identical B-cell receptor immunoglobulins and can be assigned to distinct stereotyped subsets. Although preliminary evidence suggests that B-cell receptor immunoglobulin stereotypy is relevant from a clinical viewpoint, this aspect has never been explored in a systematic manner or in a cohort of adequate size that would enable clinical conclusions to be drawn. METHODS: For this retrospective, multicentre study, we analysed 8593 patients with CLL for whom immunogenetic data were available. These patients were followed up in 15 academic institutions throughout Europe (in Czech Republic, Denmark, France, Greece, Italy, Netherlands, Sweden, and the UK) and the USA, and data were collected between June 1, 2012, and June 7, 2013. We retrospectively assessed the clinical implications of CLL B-cell receptor immunoglobulin stereotypy, with a particular focus on 14 major stereotyped subsets comprising cases expressing unmutated (U-CLL) or mutated (M-CLL) immunoglobulin heavy chain variable genes. The primary outcome of our analysis was time to first treatment, defined as the time between diagnosis and date of first treatment. FINDINGS: 2878 patients were assigned to a stereotyped subset, of which 1122 patients belonged to one of 14 major subsets. Stereotyped subsets showed significant differences in terms of age, sex, disease burden at diagnosis, CD38 expression, and cytogenetic aberrations of prognostic significance. Patients within a specific subset generally followed the same clinical course, whereas patients in different stereotyped subsets-despite having the same immunoglobulin heavy variable gene and displaying similar immunoglobulin mutational status-showed substantially different times to first treatment. By integrating B-cell receptor immunoglobulin stereotypy (for subsets 1, 2, and 4) into the well established Döhner cytogenetic prognostic model, we showed these, which collectively account for around 7% of all cases of CLL and represent both U-CLL and M-CLL, constituted separate clinical entities, ranging from very indolent (subset 4) to aggressive disease (subsets 1 and 2). INTERPRETATION: The molecular classification of chronic lymphocytic leukaemia based on B-cell receptor immunoglobulin stereotypy improves the Döhner hierarchical model and refines prognostication beyond immunoglobulin mutational status, with potential implications for clinical decision making, especially within prospective clinical trials. FUNDING: European Union; General Secretariat for Research and Technology of Greece; AIRC; Italian Ministry of Health; AIRC Regional Project with Fondazione CARIPARO and CARIVERONA; Regione Veneto on Chronic Lymphocytic Leukemia; Nordic Cancer Union; Swedish Cancer Society; Swedish Research Council; and National Cancer Institute (NIH).