Scope and Criteria for Consideration
NAR Genomics and Bioinformatics is an interdisciplinary journal focused on genomics and bioinformatics large-scale data analysis. It aims at providing the community with high quality results, analysis and methods in all aspects of genomics and bioinformatics. Reproducibility is a strong focus of the journal, and all entries will have to comply with strict guidelines ensuring the perfect reproducibility of both experimental and bioinformatics analysis. Standard papers are expected to be compliant with NAR main guidelines and entries must fulfill the condition of scientific quality, novelty, timeliness, usefulness and usability, as established through an extensive peer-reviewing process. In the case of bioinformatics methods and analysis, usability implies a suitable implementation of the FAIR principle that requires data and software to be Findable, Accessible, Interoperable and Re-usable. NAR Genomics and Bioinformatics has a strict open-source policy and will only consider for publication contributions whose novel bioinformatics components are open source.
I. Topics of Interest
All aspects of biology dealing with genomics analysis and subsequent bioinformatics analysis and interpretation will be considered. Genomics is taken here in the broadest sense of the term and may include structure, function, evolution, editing of genomes, and in general any study dealing with the analysis of the product of a genomic activity. This includes the genome maintenance, its heritability and evolution. Manuscripts dealing with large scale analysis involving high-throughput sequencing, metabolomic, proteomic or image analysis are especially encouraged, as well as manuscripts dealing with the bioinformatics analysis of such data. Bioinformatics analysis featuring novel methods will have to be done in such a way that their absolute or relative merits in terms of accuracy can objectively quantified using relevant and suitable metrics. The editors periodically review emerging scientific areas where the journal would like to attract more submissions. Currently these include:
Omics Analysis
- Genomics
- Transcriptomics
- Hi-C
- Epigenetics
- Metabolomics
- Proteomics
Structural Biology
- Structure prediction for RNA and proteins
- Protein 3D modelling
- RNA structure modelling
- Small molecule docking prediction
- Macro-molecular complexes
Functional genomics
- Functional prediction of proteins, transcripts and regulatory genomic elements
- Evolutionary and functional analysis, including predictions
- Network analysis
- In silico re-analysis of public datasets
Single Cell Analysis
- Statistical methods for single cell analysis
- Multi scale representation
- Lineage tracing methodologies - Genetic scarring
Gene Regulation
- ChIP-Seq and transcriptomic analysis
- Ribosome profiling
- RNA mediated regulation (micro, long, etc.)
Sequence Analysis
- Read mapping
- Genome comparison
- Evolutionary based sequence alignments (pairwise and multiple)
- Motif finding
Genome/Phenome analysis
- Gene calling
- Identification of genomic structural variants
- eQTL analysis
- Genetic inference methods (GWAS, etc.)
- Image analysis methods for iQTL
- Longitudinal behavioural studies
- Image tracking
Evolutionary Analysis
- Phylogenetic Reconstruction methods
- Gene family evolution
- Population Genetics
Human Health
- Phenotypic analysis
- Methods for longitudinal analysis
- Therapeutic sequencing
- Omics based disease management
- Electronic Health Record Management
- Interoperability of human data
Animal and Plant Biology
- Non-model genome analysis
- Farmed animal genomic studies
Microbiology
- Meta-genome analysis
- Classifications methods
Databases, Benchmarking, Ontologies and Reference Resources
- Data integration
- Benchmarking methods
- Benchmarking reference datasets
- Comparative benchmarking
- Ontologies and related tools
Statistical Learning
- Clustering algorithms
- Deep learning algorithms and applications
- Methods for multidimensional data analysis (tSNE, PCA, etc…)
Information Technology
- Pipeline management methods
- High performance computation in Life Science
- Scale up solutions for Life Sciences
- Software development for Life Sciences
- Standards definition in Life Sciences
II.Categories of Publications:
Application Notes
Application notes are relatively brief communications - max 3000 words, 3 figures/tables or equivalent and 30 references. They follow the guidelines of standards papers but are expected to be focused on a single result or Methods. This format is especially encouraged to publicize a resource such as a software, a computational service, a genomic technique or a reference database.
Standard Papers
Standard papers are expected to establish a novel result within the scope of the journal. Their format and organisation are expected to follow the guidelines and requirement of NAR as detailed in Preparing Your Manuscript.
Manuscripts are expected not to go over 8,000 words with a maximum of 40 citations, with a maximum of about 6 Figures/Tables or equivalent.
Whenever these results are established using novel experimental methods, or novel bioinformatics methods, the sections of the paper dealing with these aspects are encouraged to comply with the requisite of publications dealing with such items (see other paper categories). The guidelines relating to software and reference dataset deposition in public repositories will be strictly enforced.
Novel experimental data will have to be deposited in appropriate repositories. When no such repository exists, the data will have to be deposited in the Zenodo public generic repository along with suitable meta-data.
Methods Papers
Method Papers detail methodological developments of highest originality and usefulness to the wide research community within the journal scope. These papers should report novel techniques, significant advances in existing techniques, and/or demonstrate novel utility or advantage to an extended, rather than a specialist, audience. Where appropriate, advantage to existing techniques should be demonstrated at a comparative level and/or proven effective at a global, genome-wide scale. All Methods papers are subject to the same requirements as standard articles regarding availability of research materials and computational executables and/or source codes, as described for individual categories below and in 'Journal Policies'.
Method papers include Bioinformatics methods paper that should detail computational methodological developments of the highest originality and usefulness. They must relate to the topics of interests and are expected to be of immediate usage to a significant fraction of NAR Genomics and Bioinformatics readership.
Methods must be able to deal with large scale data analysis with demonstrated suitability for large scale computational architectures (HPC, Clouds). The methods must deliver results whose biological merits can be quantified and/or compared with alternative protocols.
Submissions will only be considered if they meet the basic level of requirements for scientific quality and FAIR usage (i.e. Findable, Accessible, Interoperable and Re-usable) as detailed in the Reference Dataset and Software requirement section. Method papers in their most complete form may feature one or more of the three following components:
- a method, expected to produce an output whose biological merits are quantifiable
- a benchmark methodology providing a metrics allowing the above mention quantification.
- a reference dataset that may be needed to run the benchmark.
A bioinformatics paper may contain any of these three components, provided third party resources exists that may replace the missing components (e.g. one may submit a new reference dataset used to benchmark publicly available methods using established metrics).
Each of these components will have to individually comply with the requirements for data and software submission.
Submissions may involve completely new methods, as well as new applications and combinations of existing technologies (e.g. new pipelines). In such cases, the authors are expected to establish why the dissemination of these new applications constitutes a clear improvement for the community over existing alternatives. For instance, entries describing the parameter space exploration of an existing method will only be considered if this exploration offers novel scientific insights onto the meaning of these parameters, or if it comes along with novel benchmarking methodologies equally well justified.
Submissions will have to include benchmarks against an adequate number of relevant alternative state of the art technologies. The benchmarks will have to be suitable to the problem, scientifically sound, explicit and reflect statistically significant levels of improvement. Theses benchmarks must include an evaluation methodology and, whenever applicable a reference dataset. In cases when such methodologies and datasets are already available from third parties, the authors are also expected to provide benchmarks on these resources so as to compare them with their novel benchmarking methodology. Novel benchmark methods will be separately evaluated according to the “Benchmark Papers” guidelines while novel reference datasets will also be separately evaluated following the “Reference Dataset Paper” criteria. Note that manuscript reporting novel methods while only relying on existing benchmarks and reference datasets are totally acceptable.
Upon review the referees may only recommend partial acceptation excluding some of the submitted components (e.g. a method may be recommended for publication on the basis of its validation on available benchmarks, but novel benchmarks proposed along with the method may not be accepted). Submissions will therefore have to be written in a highly structured way so as to allow rapid manuscript re-organisation. In practice this will mean separate and explicit section for each novel method, dataset and benchmark metrics.
Methods and Benchmark Surveys
Survey article are expected to provide an exhaustive combination of benchmarks using existing methods, benchmark strategies and datasets. Prior to submission the authors are encouraged to contact the editor to discuss the project. In general the authors are expected to have a strong track record in the field of the survey.
Such surveys will only be considered if all the benchmarked components meet the requirements as outlined in the Reference Dataset and Software requirement section. The journal will not publish any benchmark results on non open source software. It is the responsibility of the authors to secure the required permissions with third parties whenever these may be needed.
In order to be considered for publication, benchmark survey must come along with a review of all considered benchmark component as well as an in-depth discussion of the global benchmark results. Whenever possible such discussion should highlight the suitability of benchmarks in decision making process when selecting a computational protocol
Benchmark summary authors should have a track record of publication within the field covered by the article.
Opinion articles
Opinion articles are original scientific contribution that offer a new point of view on a specific topic within NAR Genomics and Bioinformatics scope. They may include white papers, comments on existing literature and scientific opinions of various nature. These contributions will undergo formal scientific review and are expected to meet the standards of any review. Conclusions and suggestions must be supported by a sound argumentation and by the appropriate reference literature. Authors are invited to contact the Editor-in-Chief for any project they may have that would fit within this category.
Reference Dataset and Software requirement
NARGAB is meant to provide an entry point to high quality software and reference datasets. Special attention will therefore be given at making sure the entries meet a certain number of requirements, in line with the FAIR principle (i.e. Findable, Accessible, Interoperable, Re-usable).
In order to achieves these various aims both data and methods are expected to be submitted in a very specific fashion that follows the FAIR principle as outlined below:
Findable.
Data. Whenever applicable, data will be submitted to dedicated repositories, and will be referred to in the paper by the provided accession number. When no dedicated repository is available the authors are expected to submit data on Zenodo (https://zenodo.org). Data must be submitted along with sufficient documentation and the links must be provided in the paper.
Software. Must be submitted in GitHub (https://github.com/). Whenever the deployment of the software involves compilation of binaries, a docker container must be submitted to the BioContainers section of GitHub (https://github.com/BioContainers).
Accessible.
The use of public repositories is meant to insure long-term accessibility. Storage in the local resources of the authors will not be accepted. Note that in the case of a public web service offering computation, the authors must guarantee two years of availability from the day the paper goes on print and access must be allowed without registration requirements.
Interoperable.
The software and datasets will come along with a precise and exhaustive description of the supported formats (input, output, references). Use of adequate ontologies such as EDAM is strongly encouraged whenever relevant to the software or data being considered.
Re-usable.
Manuscripts will only be considered for review if the new software is released as open-source and include a license recognised by the Open Source Initiative (OSI) foundation. Software will have to come along with a simple commands (electronic lab notebook) allowing its deployment on the reference datasets.
