About this Research Topic
Imaging and omics data science have together taken center stage in recent years by addressing a wide array of biological problems, providing high-resolution views of various biological processes. Despite the recent methodological advances, optimal analysis of imaging and omics data to extract meaningful biological results remains challenging, not only due to high dimensionality but also due to the presence of hierarchical, spatial, and temporal dependency both within and across modalities.
Both these technologies (omics and imaging), independently led to the complementary development of novel computational and statistical methods that encompass data preprocessing, modeling, and inference. Despite the progress, there is still much work to be done to meet the challenges and make use of the opportunities posed by the resulting new data types.
Although there are special issues in statistical and computational biology journals focusing on each individual data analysis, researchers from these fields as well as practitioners of these technologies would greatly benefit from a combined issue, where it is possible to exchange ideas, raise new questions, and form future collaborations, building upon the lessons learned in one field and reverse-translating the gained knowledge from one field to the other. This is particularly relevant as both these technologies generate data that have similar downstream characteristics with important differences: they are typically noisy, multimodal, heterogeneous, sparse, with confounding effects unique to each individual layer, exhibiting substantial feature- and platform-specific technological variability both within and across layers. Due to these apparent similarities and differences, many of the methods developed in the omics field have been successfully transported to the imaging field (with necessary modifications) and vice versa, and in recent years many researchers have made the seamless transition from one field to the other.
The motivation for a focused topic on imaging and omics data science is multi-fold. We believe this topic is both timely and significant for improving our current understanding of omics and imaging technologies and their implications in drug discovery and development. The diverse collection of papers on this topic will thus (i) provide a useful reference for both current and future investigators in translational and clinical research allowing sufficient cross-talk between these two paradigm-shifting fields, and (ii) establish best practice guidelines for analyzing and integrating imaging and omics data.
As a continuation of the successful first and second Research Topics, through this special issue we aim to bring together the best of both worlds by combining cutting-edge technological and computational advancements in each of these respective fields, to (i) identify new challenges in data analysis and modeling, (ii) provide a platform for interdisciplinary dialogue, and (iii) help shape future directions for these burgeoning fields, including but not limited to:
• Novel statistical and computational approaches including but not limited to downstream and upstream analytical methods for analyzing (potentially multimodal) genomics, metagenomics, metabolomics, proteomics, single-cell, and spatial transcriptomics datasets
• Novel statistical and computational approaches including but not limited to downstream and upstream analytical methods for analyzing (potentially multimodal) digital pathology, microscopy, radiology, and other imaging-based data
• Spatial, temporal (longitudinal), and spatiotemporal modeling of the omics and imaging data, possibly in combination with other non-omics and non-imaging data layers
• Integrative/mapping approaches for precision medicine potentially combining multiple omics and/or imaging data types across samples or studies, and quality control of individual data types necessary for successful data integration/mapping
• General domain-agnostic methods and visualization techniques for normalization, batch effect correction, differential analysis, variable selection, classification, or clustering analysis of multi-platform genomics and imaging data with convincing real data applications
Topic Editor Himel Malick and Suvo Chatterjee are employed by Merck Sharp & Dohme Corp. All other Topic Editors declare no competing interests with regard to the Research Topic subject.
Both these technologies (omics and imaging), independently led to the complementary development of novel computational and statistical methods that encompass data preprocessing, modeling, and inference. Despite the progress, there is still much work to be done to meet the challenges and make use of the opportunities posed by the resulting new data types.
Although there are special issues in statistical and computational biology journals focusing on each individual data analysis, researchers from these fields as well as practitioners of these technologies would greatly benefit from a combined issue, where it is possible to exchange ideas, raise new questions, and form future collaborations, building upon the lessons learned in one field and reverse-translating the gained knowledge from one field to the other. This is particularly relevant as both these technologies generate data that have similar downstream characteristics with important differences: they are typically noisy, multimodal, heterogeneous, sparse, with confounding effects unique to each individual layer, exhibiting substantial feature- and platform-specific technological variability both within and across layers. Due to these apparent similarities and differences, many of the methods developed in the omics field have been successfully transported to the imaging field (with necessary modifications) and vice versa, and in recent years many researchers have made the seamless transition from one field to the other.
The motivation for a focused topic on imaging and omics data science is multi-fold. We believe this topic is both timely and significant for improving our current understanding of omics and imaging technologies and their implications in drug discovery and development. The diverse collection of papers on this topic will thus (i) provide a useful reference for both current and future investigators in translational and clinical research allowing sufficient cross-talk between these two paradigm-shifting fields, and (ii) establish best practice guidelines for analyzing and integrating imaging and omics data.
As a continuation of the successful first and second Research Topics, through this special issue we aim to bring together the best of both worlds by combining cutting-edge technological and computational advancements in each of these respective fields, to (i) identify new challenges in data analysis and modeling, (ii) provide a platform for interdisciplinary dialogue, and (iii) help shape future directions for these burgeoning fields, including but not limited to:
• Novel statistical and computational approaches including but not limited to downstream and upstream analytical methods for analyzing (potentially multimodal) genomics, metagenomics, metabolomics, proteomics, single-cell, and spatial transcriptomics datasets
• Novel statistical and computational approaches including but not limited to downstream and upstream analytical methods for analyzing (potentially multimodal) digital pathology, microscopy, radiology, and other imaging-based data
• Spatial, temporal (longitudinal), and spatiotemporal modeling of the omics and imaging data, possibly in combination with other non-omics and non-imaging data layers
• Integrative/mapping approaches for precision medicine potentially combining multiple omics and/or imaging data types across samples or studies, and quality control of individual data types necessary for successful data integration/mapping
• General domain-agnostic methods and visualization techniques for normalization, batch effect correction, differential analysis, variable selection, classification, or clustering analysis of multi-platform genomics and imaging data with convincing real data applications
Topic Editor Himel Malick and Suvo Chatterjee are employed by Merck Sharp & Dohme Corp. All other Topic Editors declare no competing interests with regard to the Research Topic subject.
Keywords: Genomics, Metagenomics, Microbiome, Metatranscriptomics, Metabolomics, Proteomics, scRNASeq, RNASeq, Omics, Multiomics, Spatial Transcriptomics, Data Science, Digital Pathology, Imaging Genomics, Imaging, Data Integration, Bioinformatics, Biostatistics, Microscopy, CODEX, Spatial Metabolomics, Radiology, Computational Biology, Statistical Genomics, Machine Learning, AI
Important Note: All contributions to this Research Topic must be within the scope of the section and journal to which they are submitted, as defined in their mission statements. Frontiers reserves the right to guide an out-of-scope manuscript to a more suitable section or journal at any stage of peer review.
