This special issue of Bioinformatics serves as the proceedings of the 31st annual conference on Intelligent Systems for Molecular Biology (ISMB) co-organized with the 22nd European Conference on Computational Biology (ECCB). This conference took place on 23–27 July 2023, in Lyon, France. ISMB is the leading international forum for presenting new research results, disseminating methods and techniques, and facilitating discussions among leading researchers, practitioners, and students in the field. In addition, ISMB is the flagship conference of the International Society for Computational Biology (ISCB). Joining forces with ECCB, ISMB/ECCB2023 is the largest event of the computational biology community. Due to ongoing issues related to the COVID-19 pandemic, the ISMB/ECCB 2023 meeting was run as a hybrid conference, with online participants from all around the world complementing the on-site participants. The papers published in this volume were selected from 336 submitted full length papers, featuring original research. The submitted papers were thoroughly reviewed with each paper receiving 4.04 reviews on average, for a total of 1358 reports. For the review purpose, the submitted manuscripts were assigned to one of 11 scientific areas according to authors’ preference and research topic, allowing for minor adjustments to avoid conflicts of interest. In addition to selecting an area, the authors could also designate a particular Community of Special Interest (COSI; Table 1) that would provide the best forum for presentation of their paper. The research areas covered a broad spectrum of topics (Table 2) and also included a special General Computational Biology area intended for submissions on emerging topics or for those manuscripts that did not fit well in other reviewing areas. COSI distribution of accepted ISMB 2023 Proceedings papers. Thematic areas of ISMB/ECCB 2023 proceedings talks. The table lists the Area Chairs for each theme, the number of reviewed papers, the number of accepted papers, and the acceptance rate for each area. The reviewing of the submissions was overseen by the Senior Program Committee (SPC), which included the Proceedings Chairs (this editorial’s authors) and Area Chairs (AC, listed in Table 2). Several of the ACs were nominated by COSIs or by the ISMB Steering Committee. Members of the SPC were responsible for recruiting the Program Committee. Reviewers (Program Committee Members and subreviewers) judged the papers based on the novelty of computational approaches, relevance of biological questions, importance of biological insights, clarity of presentations, expected impact, compatibility with the conference format and, most importantly, correctness, completeness, and reproducibility of the studies. After submitting their reviews, the reviewers had the opportunity to discuss the papers and refine the scores. The ACs facilitated these discussions, oversaw the review process and made sure reviews were detailed and consistent with the overall decision. Final acceptance decisions were made in agreement with the entire SPC. Throughout the reviewing process, we followed a stringent policy of guarding conflicts of interest. Submissions that had any association to an Area Chair were reassigned to a different area. Care was taken not to assign papers to reviewers with a perceived conflict either (e.g. co-authorship or same institution). The definition of conflict was defined as broadly as possible—including collaborators (present and past few years), same institution (present, past few years, planned future moves), family relations, advisees/advisors, as well as any personal conflicts that could cause the appearance of conflict of interest, or that could genuinely interfere with objective reviewing. Finally, as Proceedings Chairs, we refrained from submitting papers to the conference. Among the 336 submissions, 60 were accepted for presentation at ISMB/ECCB 2023 and publication in the proceedings, conditioned on revisions having properly addressed the comments of the reviewers. In a few cases, the authors had to be reminded to release their source code alongside their manuscript, or clarify some aspects of their evaluations. This year, all conditionally accepted papers were revised and subsequently judged to have properly addressed the concerns of the reviewers. They were subsequently accepted for the conference proceedings, resulting in a 17.9% acceptance rate overall. The acceptance rates for individual Areas are shown in Table 2. Accepted papers were assigned to COSIs based on the preferences of both authors and COSI organizers (Table 1). We are deeply grateful to the Area Chairs, the 317 members of the Program Committee and the 368 subreviewers for their outstanding efforts in conducting thorough and timely reviews. Their contribution is at the core of the scientific quality of the conference. We also thank Steven Leard, Diane Kovats, and Seth Munholland for their support, guidance, and handling logistical questions and all the other members of the ISMB Steering Committee for their expert advise and supervision. We also thank the team at Oxford University Press for producing this special proceedings volume. We finally thank all the authors for submitting their work. These proceedings would not be possible without the scientific ingenuity of the contributors of all the papers. We recognize that, despite our best efforts, the selection process is necessarily imperfect, and some outstanding work will have been missed. Nonetheless, we hope that all authors received helpful feedback on their work. Finally, we want to thank all the keynote speakers, presenters, and all conference participants. Thank you all for making this meeting possible and the entire ISMB/ECCB community to continue to thrive. None declared. S.R. was supported by the James McDonnell Foundation Scholar award (220020473). Y.P. was supported by French Agence Nationale de la Recherche through the Decrypted (ANR-19-CE30-0021), PaRNAssus (ANR-19-CE45-0023), and INSSANE (ANR-21-CE45-0034) grants.

Understanding the impact of regulatory variants on complex phenotypes is a significant challenge because the genes and pathways that are targeted by such variants and the cell type context in which regulatory variants operate are typically unknown. Cell-type-specific long-range regulatory interactions that occur between a distal regulatory sequence and a gene offer a powerful framework for examining the impact of regulatory variants on complex phenotypes. However, high-resolution maps of such long-range interactions are available only for a handful of cell types. Furthermore, identifying specific gene subnetworks or pathways that are targeted by a set of variants is a significant challenge. We have developed L-HiC-Reg, a Random Forests regression method to predict high-resolution contact counts in new cell types, and a network-based framework to identify candidate cell-type-specific gene networks targeted by a set of variants from a genome-wide association study (GWAS). We applied our approach to predict interactions in 55 Roadmap Epigenomics Mapping Consortium cell types, which we used to interpret regulatory single nucleotide polymorphisms (SNPs) in the NHGRI-EBI GWAS catalogue. Using our approach, we performed an in-depth characterization of fifteen different phenotypes including schizophrenia, coronary artery disease (CAD) and Crohn's disease. We found differentially wired subnetworks consisting of known as well as novel gene targets of regulatory SNPs. Taken together, our compendium of interactions and the associated network-based analysis pipeline leverages long-range regulatory interactions to examine the context-specific impact of regulatory variation in complex phenotypes.

Graphs provide ubiquitous powerful representations of interacting components in many complex systems. In molecular biology, graphs are used to represent interactions between molecular components; for example, protein-protein interactions represent interactions between proteins, while gene regulatory networks describe interactions between transcription factor proteins and genes. A common property of graphs for complex systems, including molecular networks, is the presence of community structures or modules, defined by groups of densely interacting nodes. In protein-protein interaction networks, such modules often correspond to key biological processes needed for cells to accomplish different tasks [1]. Disruptions to module connectivity have been implicated in many diseases [13]. Therefore, the accurate detection of these modules is a key problem in the analysis of molecular networks [14]. Graph representation learning [15], which aims to embed graph nodes in a d-dimensional space (Fig 1), offers a powerful framework for performing machine learning tasks on graphs, including clustering. Both deep and shallow graph representation algorithms exist; however, deep representation learning methods have been designed and tested on social networks, citation networks and recommendation systems. Therefore, it is unclear if deep representation learning methods are beneficial for module detection of molecular networks. In this work, we compare several state-of-the-art shallow and deep learning frameworks on both real molecular networks and synthetic benchmark networks.

This special issue of Bioinformatics serves as the proceedings of the 30th annual conference on Intelligent Systems for Molecular Biology (ISMB), which took place on July 10–14, 2022, in Madison, WI, USA. ISMB is the leading international forum for presenting new research results, disseminating methods and techniques and facilitating discussions among leading researchers, practitioners and students in the field. In addition, ISMB is the flagship conference of the International Society for Computational Biology. Due to the worldwide COVID-19 pandemic, the ISMB 2022 meeting was run as a hybrid conference, with online participants from all around the world complementing the on-site participants. The papers published in this volume were selected from 243 submitted full-length papers featuring original research. The submitted papers were thoroughly reviewed with each paper receiving 4.08 reviews on average. For the review purpose, the submitted manuscripts were assigned to one of the 11 scientific areas according to the authors’ preference and research topic, allowing for minor adjustments to avoid conflicts of interest. In addition to selecting one of the 11 areas, the authors could also designate a particular Community of Special Interest (COSI; Table 1) that would provide the best forum for the presentation of their paper. The 11 research areas covered a broad spectrum of topics (Table 2) and also included a special General Computational Biology area intended for submissions on emerging topics or for those manuscripts that did not fit well in other reviewing areas. This year we also sought papers in a new area, equity-focused research. COSI distribution of accepted ISMB 2022 proceedings papers COSI distribution of accepted ISMB 2022 proceedings papers Thematic areas of ISMB 2022 Note: The table lists the Area Chairs for each theme, the number of reviewed papers, the number of accepted papers and the acceptance rate for each area. Thematic areas of ISMB 2022 Note: The table lists the Area Chairs for each theme, the number of reviewed papers, the number of accepted papers and the acceptance rate for each area. The reviewing of the submissions was overseen by the Senior Program Committee (SPC), which included the Proceedings Chairs (this editorial’s authors) and Area Chairs (AC, listed in Table 2). Several of the ACs were nominated by COSIs or by the ISMB Steering Committee. Members of the SPC were responsible for recruiting the Program Committee. Reviewers (Program Committee Members and sub-reviewers recruited by them) judged the papers based on the novelty of computational approaches, the relevance of biological questions, the importance of biological insights, clarity of presentation, correctness and completeness of the study and expected impact. After submitting their reviews, the reviewers had the opportunity to discuss the papers and refine the scores. The ACs facilitated these discussions, oversaw the review process and made sure reviews were detailed and consistent with the overall decision. Final acceptance decisions were made by the entire SPC. Throughout the reviewing process, we followed a stringent policy of guarding conflicts of interest. Submissions that had any association with an Area Chair were reassigned to a different area. Care was taken not to assign papers to reviewers with a perceived conflict either (e.g. co-authorship or same institution). The definition of conflict was defined as broadly as possible—including collaborators (present and past few years), same institution (present, past few years and planned future moves), family relations, advisees/advisors, as well as any personal conflicts that could cause the appearance of a conflict of interest, or that could genuinely interfere with objective reviewing. Finally, as Proceedings Chairs, we refrained from submitting papers to the conference. Among the 243 submissions, 48 were accepted for presentation at ISMB 2022 and publication in the proceedings, conditioned on revisions properly addressing the comments of the reviewers. In a few cases, the authors had to be reminded to release their source code alongside their manuscript. This year, all 48 conditionally accepted papers were revised and subsequently judged to have properly addressed the concerns of the reviewers and were accepted for the conference proceedings, resulting in a 19.8% acceptance rate overall. The acceptance rates for individual areas are shown in Table 2. Accepted papers were assigned to COSIs based on the preferences of both authors and COSI organizers (Table 1). We are deeply grateful to the Area Chairs, the 304 members of the Program Committee and the 270 sub-reviewers for their outstanding efforts in conducting thorough and timely reviews. Their contribution is at the core of the scientific quality of the conference. We also thank Steven Leard, Diane Kovats and Seth Munholland for their support, guidance and handling logistical questions and all the other members of the ISMB Steering Committee for their expert advice and supervision. We also thank the team at Oxford University Press for producing this special proceedings volume. We also thank all the authors for submitting their work. These proceedings would not be possible without the scientific ingenuity of the contributors of all the papers. We recognize that, despite our best efforts, the selection process is necessarily imperfect, and some outstanding work will have been missed. Nonetheless, we hope that all authors received helpful feedback on their work. Finally, we want to thank all the keynote speakers, presenters, and all conference participants. Thank you all for making this meeting possible and the entire ISMB community to continue to thrive. No new data were generated or analysed in support of this research. This work was supported by the James McDonnell Foundation to S.R.; Swiss National Science Foundation grants (205085 and 186397 to C.D.). Conflict of Interest: none declared.

MOTIVATION: Identifying the shared and pathogen-specific components of host transcriptional regulatory programs is important for understanding the principles of regulation of immune response. Recent efforts in systems biology studies of infectious diseases have resulted in a large collection of datasets measuring host transcriptional response to various pathogens. Computational methods to identify and compare gene expression modules across different infections offer a powerful way to identify strain-specific and shared components of the regulatory program. An important challenge is to identify statistically robust gene expression modules as well as to reliably detect genes that change their module memberships between infections. RESULTS: We present MULCCH (MULti-task spectral Consensus Clustering for Hierarchically related tasks), a consensus extension of a multi-task clustering algorithm to infer high-confidence strain-specific host response modules under infections from multiple virus strains. On simulated data, MULCCH more accurately identifies genes exhibiting pathogen-specific patterns compared to non-consensus and nonmulti-task clustering approaches. Application of MULCCH to mammalian transcriptional response to a panel of influenza viruses showed that our method identifies clusters with greater coherence compared to non-consensus methods. Further, MULCCH derived clusters are enriched for several immune system-related processes and regulators. In summary, MULCCH provides a reliable module-based approach to identify molecular pathways and gene sets characterizing commonality and specificity of host response to viruses of different pathogenicities. AVAILABILITY AND IMPLEMENTATION: The source code is available at https://bitbucket.org/roygroup/mulcch CONTACT: [email protected] SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

Mammalian host response to pathogenic infections is controlled by a complex regulatory network connecting regulatory proteins such as transcription factors and signaling proteins to target genes. An important challenge in infectious disease research is to understand molecular similarities and differences in mammalian host response to diverse sets of pathogens. Recently, systems biology studies have produced rich collections of omic profiles measuring host response to infectious agents such as influenza viruses at multiple levels. To gain a comprehensive understanding of the regulatory network driving host response to multiple infectious agents, we integrated host transcriptomes and proteomes using a network-based approach. Our approach combines expression-based regulatory network inference, structured-sparsity based regression, and network information flow to infer putative physical regulatory programs for expression modules. We applied our approach to identify regulatory networks, modules and subnetworks that drive host response to multiple influenza infections. The inferred regulatory network and modules are significantly enriched for known pathways of immune response and implicate apoptosis, splicing, and interferon signaling processes in the differential response of viral infections of different pathogenicities. We used the learned network to prioritize regulators and study virus and time-point specific networks. RNAi-based knockdown of predicted regulators had significant impact on viral replication and include several previously unknown regulators. Taken together, our integrated analysis identified novel module level patterns that capture strain and pathogenicity-specific patterns of expression and helped identify important regulators of host response to influenza infection.

Regulatory networks that control gene expression are important in diverse biological contexts including stress response and development. Each gene's regulatory program is determined by module-level regulation (e.g. co-regulation via the same signaling system), as well as gene-specific determinants that can fine-tune expression. We present a novel approach, Modular regulatory network learning with per gene information (MERLIN), that infers regulatory programs for individual genes while probabilistically constraining these programs to reveal module-level organization of regulatory networks. Using edge-, regulator- and module-based comparisons of simulated networks of known ground truth, we find MERLIN reconstructs regulatory programs of individual genes as well or better than existing approaches of network reconstruction, while additionally identifying modular organization of the regulatory networks. We use MERLIN to dissect global transcriptional behavior in two biological contexts: yeast stress response and human embryonic stem cell differentiation. Regulatory modules inferred by MERLIN capture co-regulatory relationships between signaling proteins and downstream transcription factors thereby revealing the upstream signaling systems controlling transcriptional responses. The inferred networks are enriched for regulators with genetic or physical interactions, supporting the inference, and identify modules of functionally related genes bound by the same transcriptional regulators. Our method combines the strengths of per-gene and per-module methods to reveal new insights into transcriptional regulation in stress and development.

MOTIVATION: Condition-specific networks capture system-wide behavior under varying conditions such as environmental stresses, cell types or tissues. These networks frequently comprise parts that are unique to each condition, and parts that are shared among related conditions. Existing approaches for learning condition-specific networks typically identify either only differences or only similarities across conditions. Most of these approaches first learn networks per condition independently, and then identify similarities and differences in a post-learning step. Such approaches do not exploit the shared information across conditions during network learning. RESULTS: We describe an approach for learning condition-specific networks that identifies the shared and unique subgraphs during network learning simultaneously, rather than as a post-processing step. Our approach learns networks across condition sets, shares data from different conditions and produces high-quality networks that capture biologically meaningful information. On simulated data, our approach outperformed an existing approach that learns networks independently for each condition, especially for small training datasets. On microarray data of hundreds of deletion mutants in two, yeast stationary-phase cell populations, the inferred network structure identified several common and population-specific effects of these deletion mutants and several high-confidence cases of double-deletion pairs, which can be experimentally tested. Our results are consistent with and extend the existing knowledge base of differentiated cell populations in yeast stationary phase. AVAILABILITY AND IMPLEMENTATION: C++ code can be accessed from http://www.broadinstitute.org/~sroy/condspec/ .

In many real-world domains, undirected graphical models such as Markov random fields provide a more natural representation of the statistical dependency structure than directed graphical models. Unfortunately, structure learning of undirected graphs using likelihood-based scores remains difficult because of the intractability of computing the partition function. We describe a new Markov random field structure learning algorithm, motivated by canonical parameterization of Abbeel et al. We provide computational improvements on their parameterization by learning per-variable canonical factors, which makes our algorithm suitable for domains with hundreds of nodes. We compare our algorithm against several algorithms for learning undirected and directed models on simulated and real datasets from biology. Our algorithm frequently outperforms existing algorithms, producing higher-quality structures, suggesting that enforcing consistency during structure learning is beneficial for learning undirected graphs.

UNLABELLED: We have developed a new software system, REgulatory Network generator with COmbinatorial control (RENCO), for automatic generation of differential equations describing pre-transcriptional combinatorics in artificial regulatory networks. RENCO has the following benefits: (a) it explicitly models protein-protein interactions among transcription factors, (b) it captures combinatorial control of transcription factors on target genes and (c) it produces output in Systems Biology Markup Language (SBML) format, which allows these equations to be directly imported into existing simulators. Explicit modeling of the protein interactions allows RENCO to incorporate greater mechanistic detail of the transcription machinery compared to existing models and can provide a better assessment of algorithms for regulatory network inference. AVAILABILITY: RENCO is a C++ command line program, available at http://sourceforge.net/projects/renco/

BACKGROUND: Modeling of gene expression data from time course experiments often involves the use of linear models such as those obtained from principal component analysis (PCA), independent component analysis (ICA), or other methods. Such methods do not generally yield factors with a clear biological interpretation. Moreover, implicit assumptions about the measurement errors often limit the application of these methods to log-transformed data, destroying linear structure in the untransformed expression data. RESULTS: In this work, a method for the linear decomposition of gene expression data by multivariate curve resolution (MCR) is introduced. The MCR method is based on an alternating least-squares (ALS) algorithm implemented with a weighted least squares approach. The new method, MCR-WALS, extracts a small number of basis functions from untransformed microarray data using only non-negativity constraints. Measurement error information can be incorporated into the modeling process and missing data can be imputed. The utility of the method is demonstrated through its application to yeast cell cycle data. CONCLUSION: Profiles extracted by MCR-WALS exhibit a strong correlation with cell cycle-associated genes, but also suggest new insights into the regulation of those genes. The unique features of the MCR-WALS algorithm are its freedom from assumptions about the underlying linear model other than the non-negativity of gene expression, its ability to analyze non-log-transformed data, and its use of measurement error information to obtain a weighted model and accommodate missing measurements.

SUMMARY: Using replicated human serum samples, we applied an error model for proteomic differential expression profiling for a high-resolution liquid chromatography-mass spectrometry (LC-MS) platform. The detailed noise analysis presented here uses an experimental design that separates variance caused by sample preparation from variance due to analytical equipment. An analytic approach based on a two-component error model was applied, and in combination with an existing data driven technique that utilizes local sample averaging, we characterized and quantified the noise variance as a function of mean peak intensity. The results indicate that for processed LC-MS data a constant coefficient of variation is dominant for high intensities, whereas a model for low intensities explains Poisson-like variations. This result leads to a quadratic variance model which is used for the estimation of sample preparation noise present in LC-MS data.