Reproducible, portable, and efficient ancient genome reconstruction with nf-core/eager

This manuscript (permalink) was automatically generated from apeltzer/eager2-paper@91f3ed6 on June 11, 2020.

Authors

Abstract

The broadening utilisation of ancient DNA (aDNA) to address archaeological, palaeontological, and biological questions is resulting in a rising diversity in the size of laboratories and scale of analyses being performed. In the context of this heterogeneous landscape, we present nf-core/eager, an advanced and entirely redesigned pipeline for the analysis of ancient genomic data. nf-core/eager builds on existing ideas and concepts introduced in the original EAGER pipeline, and improves various aspects of the analysis procedure by employing computational frameworks such as Nextflow and nf-core. The pipeline aims to address three main themes: accessibility and adaptability to different research groups and their computing configurations, reproducibility to ensure robust analytical standards in the field, and updating the EAGER pipeline to the latest routine ancient genomic practises. This new version of EAGER has been developed within the nf-core initiative, to ensure high quality software development and maintenance support; contributing to a long-term lifecycle for the pipeline. nf-core/eager will assist in ensuring that ancient DNA sequencing data can be used by a diverse range of research groups and fields.

Introduction

Ancient DNA (aDNA) has become a widely accepted source of biological data, helping to provide new perspective for a range of fields including archaeology, ecology, cultural heritage, and palaeontology. The utilisation of next-generation sequencing has allowed the recovery of aDNA from a wide variety of sources, including but not limited to, the skeletal remains of animals [1,2,3,4], modern and archaic humans [5,6,7,8], bacteria [9,10,11], viruses [12,13], plants [14,15], palaeofaeces [16,17], dental calculus [18,19], sediments [20,21], medical slides [22], parchment [23], and most recently, ancient ‘chewing gum’ [24,25]. Improvement in laboratory protocols to increase yields of otherwise trace amounts of DNA has at the same time led to studies that can total hundreds of ancient individuals [26,27], spanning single [28] to thousands of organisms [18]. These differences of disciplines have led to a heterogeneous landscape in terms of the types of analyses undertaken, and their computational resource requirements [29,30]. Taking into consideration the unequal distribution of resources (and infrastructure such as internet connection), streamlined and efficient pipelines can help increase accessibility to high-quality analyses.

The degraded nature of aDNA poses an extra layer of complexity to standard modern genomic analysis. Through a variety of processes [31], DNA molecules fragment over time, resulting in ultra-short molecules [32]. These sequences have low nucleotide complexity making it difficult to identify with precision which part of the genome a read is derived from. When fragmentation is not a ‘clean break’, this can lead to uneven ends with single-stranded ‘overhangs’ at end of molecules, which are susceptible to chemical processes such as deamination that lead to misincorporation of bases during library construction [33]. On top of this, taphonomic processes such as heat, moisture, and microbial- and burial-environment processes lead to varying rates of degradation [34,35], where the original DNA content of a sample is lost and supplanted by modern environmental DNA. Later handling by archaeologists, museum curators, and scientists can also contribute ‘modern’ contamination. While these characteristics can help provide evidence towards the ‘authenticity’ of true aDNA sequences (e.g. aDNA C>T ‘damage’ profiles [36]), they also pose specific challenges such as unspecific DNA alignment and/or low coverage and miscoding lesions that can result in low-confidence genotyping. These factors often lead to prohibitive sequencing costs to retrieve enough data for modern NGS data pipelines (such as more than 1 billion reads for a 1X depth coverage Yersinia pestis genome [37]), and thus require aDNA-tailored methods and techniques to overcome these challenges.

Two previously published and commonly used pipelines in the field are PALEOMIX [38] and EAGER [39]. These two pipelines take a similar approach to link together standard tools used for Illumina NGS data processing (sequencing quality control, sequencing adapter removal/and or paired-end read merging, mapping of reads to a reference genome, genotyping, etc.), but with a specific focus on tools that are designed for, or well-suited for aDNA (such as the bwa aln algorithm for ultra-short molecules [40] and mapDamage [41] for evaluation of aDNA characteristics). Yet, neither of these pipelines have had major updates to bring them in-line with current routine aDNA analyses. Metagenomic screening of off-target genomic reads for pathogens or microbiomes [18,19] has become particularly common, given its role in revealing widespread infectious disease and possible epidemics that had previously been undetected in the archaeological record [12,13,37,42]. Without easy access to the latest field-established analytical routines, aDNA studies risk being published without the necessary quality control checks that ensure aDNA authenticity as well as limiting the full range of possibilities from their data. Given that material from samples is limited, there are both ethical as well as economical interests to maximise analytical yield [43].

To address these shortcomings, we have completely re-implemented the latest version of the EAGER pipeline in Nextflow [44] (a domain-specific-language or ‘DSL’, specifically designed for the construction of omics analysis pipelines), introduced new features, and more flexible pipeline configuration. In addition, the newly named pipeline - nf-core/eager - has been developed in the context of the nf-core community framework [45], which enforces strict guidelines for best-practises in software development.

Results and Discussion

Scalability, Portability, and Efficiency

The re-implementation of EAGER into Nextflow offers a range of benefits over the original custom pipeline framework.

Firstly, the new framework provides immediate integration of nf-core/eager into various job schedulers in POSIX High-Performance-Cluster (HPC) environments, cloud computing resources, as well as local workstations. This portability allows users to run nf-core/eager regardless of the type of computing infrastructure or cluster size (if applicable), with minimal effort or configuration. This facilitates reproducibility and therefore maintenance of standards within the field. This is further assisted by the in-built compatibility with software environments and containers such as Conda [46], Docker [47] and Singularity [48]. These are isolated sandboxes that include all software (with exact versions) required by the pipeline, in a form that is installable and runnable by users regardless of the setup of their local software environment. Another major change with nf-core/eager is that the graphical-user-interface (GUI) set up of an EAGER run is now replaced with a command-line-interface (CLI) as the primary user interaction mode. This is more compatible and portable with most HPCs (that may not offer display of a window system), and is in line with the vast majority of bioinformatic tools. We therefore believe this will not be a hindrance to new researchers from outside computational biology. However, there are plans from Nextflow (with tower.nf [49]) and within the nf-core community to provide multiple alternatives in the near future, including a CLI wizard and a web-based input GUI.

Secondly, reproducibility is made easier through the use of ‘profiles’ that can define configuration parameters. These profiles can be managed at different hierarchical levels. HPC-level profiles can specify parameters for the computing environment (job schedulers, cache locations for containers, maximum memory and CPU resources etc.), which can be centrally managed to ensure all users of a group use the same settings. Pipeline-level profiles, specifying parameters for nf-core/eager itself, allow fast access to routine pipeline-run parameters via a single flag in the nf-core/eager run command, without having to configure each new run from scratch. Compared to the original EAGER, which utilised per-FASTQ XML files with hardcoded filepaths for a specific user’s server, nf-core/eager allows researchers to publish the specific profile used in their runs alongside their publications, to ensure other groups can generate the same results. Usage of profiles can also reduce mistakes caused by insufficient ‘prose’ based reporting of program settings that can be regularly found in the literature. The default nf-core/eager profile uses parameters evaluated in different aDNA-specific contexts (e.g. in [50]), and will be updated in each new release as new studies are published.

Finally, nf-core/eager provides improved efficiency over the original EAGER pipeline by replacing the sample-by-sample sequential processing with Nextflow’s asynchronous job parallelisation, whereby multiple pipeline steps and samples are run in parallel (in addition to natively parallelised pipeline steps). This, combined with pre-defined per-process customisation of resource parameters, reduces unnecessary resource allocation that can occur with unfamiliar users to each step of an NGS data processing pipeline. This is particularly pertinent given the increasing use of centralised HPCs or cloud computing that often use per-hour cost calculations.

Updated Workflow

nf-core/eager follows a similar structural foundation to the original version of EAGER. Given Illumina short-read FASTQ and/or BAM files and a reference FASTA file, this can be split into four main stages:

  1. Pre-processing
  2. Mapping and post-processing
  3. aDNA Evaluation and Modification
  4. Genotyping and Consensus Sequence Generation: GATK [56], VCF2Genome [39]

In nf-core/eager, all tools originally used in EAGER have been updated to their latest versions, as available on Bioconda [62] and Conda-forge [63] to ensure widespread accessibility and stability of utilised tools. The MapDamage2 (for damage profile generation) [36] and Schmutzi (for mitochondrial contamination estimation) [64] methods have not been carried over to nf-core/eager, the first because a more performant successor method is now available (DamageProfiler), and the latter because a stable release of the method could not be migrated to Bioconda. We anticipate that there will be an updated version of Schmutzi in the near future that will allow us to integrate the method again in nf-core/eager, once a version is released on Bioconda. Alternatively, However, estimation of human nuclear contamination estimation is now offered through ANGSD [65]. Support for the Bowtie2 aligner [66] following [67] will be added in the near future, after consultation with the palaeogenetics community. New tools to the basic workflow include fastp [68] for the removal of ‘poly-G’ sequencing artefacts that are common in 2-colour Illumina sequencing machines (such as the increasingly popular NextSeq and NovaSeq platforms [69]). For genotyping, we have now included FreeBayes [70] as an alternative to the human-focused GATK tools. We have also maintained the possibility of using the now unsupported GATK UnifiedGenotyper, as the GATK HaplotypeCaller performs de novo assembly around possible variants, which may not be suitable for low-coverage aDNA data.

Figure 1: Simplified schematic of the nf-core/eager workflow pipeline. Green filled bubbles indicate new functionality added over the original EAGER pipeline.

We have further extended the functionality of the pipeline by adding ancient metagenomic analysis, to identify the wider taxonomic content of a sample. We have added the possibility to screen all off-target reads (not mapped to the reference genome) with two metagenomic profilers: MALT [71,72] and Kraken2 [73]. Characterisation of properties of authentic aDNA from MALT alignments is carried out with the HOPS pipeline [74]. Ancient metagenomic studies sometimes include comparative samples from living individuals [75]. To support open data, whilst respecting personal data privacy, nf-core/eager includes a ‘strip_fastq’ script which creates raw FASTQ files, but with all reads successfully mapped to the reference genome removed. This allows safe upload of sequencing data to public repositories after removal of identifiable (human) data.

Additional functionality tailored for ancient bacterial genomics includes integration of a SNP alignment generation tool, MultiVCFAnalyzer [9], which allows assessment of cross-mapping levels from different related taxa to a reference genome - a common challenge in ancient bacterial genome reconstruction [35]. The output SNP alignment FASTA file can then be used for downstream analyses such as phylogenetic tree construction. Simple coverage statistics of particular annotations (e.g. genes) of an input reference is offered by bedtools [76], which can be used in cases such as for determining functional differences between ancient bacterial strains (as in [42]). When using a human reference genome, nf-core/eager can now also can give estimates of the biological sex of a given individual with Sex.DetERRmine [77]. A dedicated ‘endogenous DNA’ calculator (endorS.py) is also included, to provide a percentage estimate of the sequenced reads matching the reference from the total number of reads sequenced per library.

Given the large amount of sequencing often required to yield sufficient genome coverage from aDNA data, palaeogeneticists tend to use multiple (differently treated) libraries, and/or merge data from multiple sequencing runs per library. The original EAGER pipeline could only run single libraries at a time, and in these contexts required significant manual user input in merging different FASTQ or BAM files. A major upgrade in nf-core/eager is that the new pipeline supports automated processing of complex sequencing strategies for many samples. This is facilitated by the optional use as input of a simple table in TSV format, which includes file paths and additional metadata such as sample name, library name, sequencing lane, colour chemistry, and UDG treatment. This allows simultaneous processing and appropriate merging of heterogeneous data from multiple sequencing runs and/or library types.

The original EAGER pipeline required users to look through many independent output directories and files to make full assessment of their sequencing data. This has now been replaced with a much more extensive MultiQC [78] report. This tool aggregates the log files of every supported tool into a single interactive report, and assists users in making a fuller assessment of their sequencing and analysis runs. We have strived to ensure that there is a corresponding MultiQC module for every tool used by nf-core/eager where possible, to enable comprehensive evaluation of all stages of the pipeline.

An overview of the entire pipeline is shown in Fig. 1.

Accessibility

Alongside the interactive MultiQC report, we have written extensive documentation on all parts of running and interpreting the output of the pipeline. Given that a large fraction of aDNA researchers come from fields outside computational biology, and thus may have limited computational training, we have written documentation [79] that also gives guidance on how to interpret each section of the report - specifically in the context of NGS and aDNA. This includes best practices or expected output schematic images, which are published under CC-BY licenses to allow for use in other training material (an example can be seen in Fig. 2). We hope this open-access resource will make the study of aDNA more accessible to researchers new to the field, by providing practical guidelines on how to evaluate characteristics and effects of aDNA on downstream analyses.

Figure 2: Example schematic images of pipeline output documentation that can assist new users in the interpretation to next-generation-sequencing aDNA processing.

The development of nf-core/eager in Nextflow and the nf-core initiative will also improve open-source development, while ensuring the high quality of community contributions to the pipeline. While Nextflow is written primarily in Groovy, the Nextflow DSL simplifies a number of concepts to an intermediate level that bioinformaticians without Java/Groovy experience can easily access (regardless of own programming language experience). Furthermore, Nextflow places ubiquitous and more widely known command-line interfaces, such as bash, in a prominent position within the code, rather than custom java code and classes. We hope this will motivate further bug fixes and feature contributions from the community, to keep the pipeline state-of-the-art and ensure a longer life-cycle. This will also be supported by the active and welcoming nf-core community who provide general guidance and advice on developing Nextflow and nf-core pipelines.

Conclusion

nf-core/eager is an efficient, portable, and accessible pipeline for processing ancient genomic data. This re-implementation of EAGER into Nextflow and nf-core will improve reproducibility and inclusion of rapidly increasing aDNA datasets, for both large and small laboratories. Extensive documentation also enables newcomers to the field to get a practical understanding on how to interpret aDNA in the context of NGS data processing. Ultimately, nf-core/eager provides easier access to the latest tools and routine screening analyses commonly used in the field, and sets up the pipeline for remaining at the forefront of palaeogenetic analysis.

Methods

Installation

nf-core/eager requires a version of Java, Nextflow and either a functional Conda installation or Docker/Singularity container installation. A quick installation guide to follow to get started can be found in the Quickstart section of the nf-core/eager repository [80].

Running

After the installation, users can run the pipeline using standard test data by utilising some of the test profiles we provide (e.g. using Docker):

nextflow run nf-core/eager -r 2.1.0 -profile test,docker

This will download test data automatically, run the pipeline locally with all software tools containerised in a Docker image and store the output of that run in the a ‘./results’ folder of your current directory.

The default pipeline settings assumes paired end FASTQ data, and will run:

If no additional FASTA indices are given, these will also be generated.

The pipeline is highly configurable and most modules can be turned on-and-off using different flags at the request of the user, to allow high customisation to each user’s needs. For example, to include metagenomic screening of off-target reads, and sex determination based on on-target mappings of pre-clipped single-end data:

nextflow run nf-core/eager -r 2.1.0 \
-profile conda \
--input '/<path>/<to>/*/*R1*.fastq.gz' --single_end \
--fasta '/<path>/<to>/<reference>.fasta.gz' \
--skip_fastqc --skip_adapterremoval \
--run_bam_filtering --bam_discard_unmapped --bam_unmapped_type 'fastq' \
--run_metagenomic_screening --metagenomic_tool 'malt' --database '/<path>/<to>/<malt_database>' \
--run_sexdeterrmine

Profiles

In additional to private locally defined profiles, we utilise a central configuration repository to enable users from various institutions to use pipelines on their particular infrastructure more easily [81]. There are multiple resources listed in this repository with information on how to add your own configuration profile with help from the nf-core community.

Users can customise this infrastructure profile by themselves, with the nf-core community, or with their local system administrator to make sure that the pipeline runs successfully, and can then rely on the Nextflow and nf-core framework to ensure compatibility upon further infrastructure changes. For example, in order to run the nf-core/eager pipeline at the Max Planck Institute for the Science of Human History (MPI-SHH) in Jena, users only have to run:

nextflow run nf-core/eager -r 2.1.0 -profile shh_cdag,test

This runs the testing profile of the nf-core/eager pipeline with parameters specifically adapted to the HPC system at the MPI-SHH. In some cases, similar institutional configs for other institutions may already exist (originally utilised for different nf-core pipelines), so users need not write their own.

Inputs

The pipeline can be started using (raw) FASTQ files from sequencing or pre-mapped BAM files. Additionally, the pipeline requires a FASTA reference genome. If BAM input is provided, an optional conversion to FASTQ is offered, otherwise BAM files processing will start from the post-mapping stage.

If users have complex set-ups, e.g. multiple sequencing lanes that require merging of files, the pipeline can be supplied with a tab separated value (TSV) file to enable such complex data handling. Both FASTQs and BAMs can be provided in this set up. FASTQs with the same library name and sequencing chemistry but sequenced across multiple lanes will be concatenated after adapter removal and prior mapping. Libraries with the sample name and with the same UDG treatment, will be merged after deduplication. If libraries with the sample name have different UDG treatment, these will be merged after the aDNA modification stage (i.e. BAM trimming or PMDtools, if turned on), prior to genotyping, as shown in Figure 3.

Figure 3: Schematic of different processing and merging points based on the nature of different libraries, as specified in metadata of a TSV file. Dashed boxes represent optional library-specific processes

As Nextflow will automatically download files from URLs, profiles and/or TSV files can include links to publicly available data (e.g. the ENA FTP server). This assists in reproducibility, because if profiles or TSV files are uploaded with a publication, a researcher wishing to re-analyse the data in the same way can use the exact settings and merging procedures in the original publication, without having to reconstruct this from prose.

Monitoring

Users can either monitor their pipeline execution with the messages Nextflow prints to the console while running, or utilise projects such as Nextflow Tower [49] to monitor their analysis pipeline during runtime.

Output

The pipeline produces a multitude of output files in various file formats, with a more detailed listing available in the user documentation. These include metrics, statistical analysis data, and standardised output files (BAM, VCF) for close inspection and further downstream analysis, as well as a MultiQC report. If an emailing daemon is set up on the server, the latter can be emailed to users automatically, when starting the pipeline with a dedicated option (--email you@yourdomain.org).

Data and software availability

All code is available on github at https://github.com/nf-core/eager and archived with Zenodo under the DOI 10.5281/zenodo.1465061. The version of nf-core/eager that this preprint is based on is the current ‘dev’ branch of the GitHub repository (2.2.0dev), and on publication will be released as 2.2.0.

This paper was written with Manubot [82].

Competing Interests

No competing interests are declared.

Acknowledgements

We thank the nf-core community for general support and suggestions during the writing of the pipeline. We also thank Arielle Munters, Hester van Schalkwyk, Irina Velsko, Katerine Eaton, Luc Venturini, Marcel Keller, Pierre Lindenbaum, Pontus Skoglund, Raphael Eisenhofer, Torsten Günter, and Kevin Lord for bug reports and feature suggestions. We are grateful to the members of the Department of Archaeogenetics at the Max Planck Institute for the Science of Human History who performed beta testing of the pipeline. We thank the aDNA twitter community for responding to polls regarding design decisions during development.

We also want to thank Selina Carlhoff, Alexander Herbig and Wolfgang Haak for providing comments and suggestions on this manuscript, and acknowledge Christina Warinner, Stephan Schiffels and the Max Planck Society who provided funds for travel to nf-core events.

References

1. Complete Genomes Reveal Signatures of Demographic and Genetic Declines in the Woolly Mammoth
Eleftheria Palkopoulou, Swapan Mallick, Pontus Skoglund, Jacob Enk, Nadin Rohland, Heng Li, Ayça Omrak, Sergey Vartanyan, Hendrik Poinar, Anders Götherström, … Love Dalén
Current Biology (2015-05) https://doi.org/34d
DOI: 10.1016/j.cub.2015.04.007 · PMID: 25913407 · PMCID: PMC4439331

2. Recalibrating Equus evolution using the genome sequence of an early Middle Pleistocene horse
Ludovic Orlando, Aurélien Ginolhac, Guojie Zhang, Duane Froese, Anders Albrechtsen, Mathias Stiller, Mikkel Schubert, Enrico Cappellini, Bent Petersen, Ida Moltke, … Eske Willerslev
Nature (2013-06-26) https://doi.org/q7n
DOI: 10.1038/nature12323 · PMID: 23803765

3. Ancient pigs reveal a near-complete genomic turnover following their introduction to Europe
Laurent A. F. Frantz, James Haile, Audrey T. Lin, Amelie Scheu, Christina Geörg, Norbert Benecke, Michelle Alexander, Anna Linderholm, Victoria E. Mullin, Kevin G. Daly, … Greger Larson
Proceedings of the National Academy of Sciences (2019-08-27) https://doi.org/gf9hnf
DOI: 10.1073/pnas.1901169116 · PMID: 31405970 · PMCID: PMC6717267

4. Ancient DNA reveals the Arctic origin of Viking Age cod from Haithabu, Germany
Bastiaan Star, Sanne Boessenkool, Agata T. Gondek, Elena A. Nikulina, Anne Karin Hufthammer, Christophe Pampoulie, Halvor Knutsen, Carl André, Heidi M. Nistelberger, Jan Dierking, … James H. Barrett
Proceedings of the National Academy of Sciences (2017-08-22) https://doi.org/gbt8b2
DOI: 10.1073/pnas.1710186114 · PMID: 28784790 · PMCID: PMC5576834

5. 137 ancient human genomes from across the Eurasian steppes
Peter de Barros Damgaard, Nina Marchi, Simon Rasmussen, Michaël Peyrot, Gabriel Renaud, Thorfinn Korneliussen, J. Víctor Moreno-Mayar, Mikkel Winther Pedersen, Amy Goldberg, Emma Usmanova, … Eske Willerslev
Nature (2018-05-09) https://doi.org/gd8hs5
DOI: 10.1038/s41586-018-0094-2 · PMID: 29743675

6. A Draft Sequence of the Neandertal Genome
R. E. Green, J. Krause, A. W. Briggs, T. Maricic, U. Stenzel, M. Kircher, N. Patterson, H. Li, W. Zhai, M. H. Y. Fritz, … S. Paabo
Science (2010-05-06) https://doi.org/c2x
DOI: 10.1126/science.1188021 · PMID: 20448178 · PMCID: PMC5100745

7. A High-Coverage Genome Sequence from an Archaic Denisovan Individual
M. Meyer, M. Kircher, M.-T. Gansauge, H. Li, F. Racimo, S. Mallick, J. G. Schraiber, F. Jay, K. Prufer, C. de Filippo, … S. Paabo
Science (2012-08-30) https://doi.org/q8p
DOI: 10.1126/science.1224344 · PMID: 22936568 · PMCID: PMC3617501

8. The genome of the offspring of a Neanderthal mother and a Denisovan father
Viviane Slon, Fabrizio Mafessoni, Benjamin Vernot, Cesare de Filippo, Steffi Grote, Bence Viola, Mateja Hajdinjak, Stéphane Peyrégne, Sarah Nagel, Samantha Brown, … Svante Pääbo
Nature (2018-08-22) https://doi.org/cs64
DOI: 10.1038/s41586-018-0455-x · PMID: 30135579 · PMCID: PMC6130845

9. Pre-Columbian mycobacterial genomes reveal seals as a source of New World human tuberculosis
Kirsten I. Bos, Kelly M. Harkins, Alexander Herbig, Mireia Coscolla, Nico Weber, Iñaki Comas, Stephen A. Forrest, Josephine M. Bryant, Simon R. Harris, Verena J. Schuenemann, … Johannes Krause
Nature (2014-08-20) https://doi.org/f6nk4g
DOI: 10.1038/nature13591 · PMID: 25141181 · PMCID: PMC4550673

10. Integrative approach using Yersinia pestis genomes to revisit the historical landscape of plague during the Medieval Period
Amine Namouchi, Meriam Guellil, Oliver Kersten, Stephanie Hänsch, Claudio Ottoni, Boris V. Schmid, Elsa Pacciani, Luisa Quaglia, Marco Vermunt, Egil L. Bauer, … Barbara Bramanti
Proceedings of the National Academy of Sciences (2018-12-11) https://doi.org/ggfn3h
DOI: 10.1073/pnas.1812865115 · PMID: 30478041 · PMCID: PMC6294933

11. Ancient genomes reveal a high diversity of Mycobacterium leprae in medieval Europe
Verena J. Schuenemann, Charlotte Avanzi, Ben Krause-Kyora, Alexander Seitz, Alexander Herbig, Sarah Inskip, Marion Bonazzi, Ella Reiter, Christian Urban, Dorthe Dangvard Pedersen, … Johannes Krause
PLOS Pathogens (2018-05-10) https://doi.org/gdrj4v
DOI: 10.1371/journal.ppat.1006997 · PMID: 29746563 · PMCID: PMC5944922

12. Ancient hepatitis B viruses from the Bronze Age to the Medieval period
Barbara Mühlemann, Terry C. Jones, Peter de Barros Damgaard, Morten E. Allentoft, Irina Shevnina, Andrey Logvin, Emma Usmanova, Irina P. Panyushkina, Bazartseren Boldgiv, Tsevel Bazartseren, … Eske Willerslev
Nature (2018-05-09) https://doi.org/gddxvj
DOI: 10.1038/s41586-018-0097-z · PMID: 29743673

13. Neolithic and medieval virus genomes reveal complex evolution of hepatitis B
Ben Krause-Kyora, Julian Susat, Felix M Key, Denise Kühnert, Esther Bosse, Alexander Immel, Christoph Rinne, Sabin-Christin Kornell, Diego Yepes, Sören Franzenburg, … Johannes Krause
eLife (2018-05-10) https://doi.org/gdhck2
DOI: 10.7554/elife.36666 · PMID: 29745896 · PMCID: PMC6008052

14. Ancient reveals the timing and persistence of organellar genetic bottlenecks over 3,000 years of sunflower domestication and improvement
Nathan Wales, Melis Akman, Ray H. B. Watson, Fátima Sánchez Barreiro, Bruce D. Smith, Kristen J. Gremillion, M. Thomas P. Gilbert, Benjamin K. Blackman
Evolutionary Applications (2018-02-13) https://doi.org/gf568v
DOI: 10.1111/eva.12594 · PMID: 30622634 · PMCID: PMC6304678

15. The origins and adaptation of European potatoes reconstructed from historical genomes
Rafal M. Gutaker, Clemens L. Weiß, David Ellis, Noelle L. Anglin, Sandra Knapp, José Luis Fernández-Alonso, Salomé Prat, Hernán A. Burbano
Nature Ecology & Evolution (2019-06-24) https://doi.org/ggxkk8
DOI: 10.1038/s41559-019-0921-3 · PMID: 31235927

16. The Prevotella copri Complex Comprises Four Distinct Clades Underrepresented in Westernized Populations
Adrian Tett, Kun D. Huang, Francesco Asnicar, Hannah Fehlner-Peach, Edoardo Pasolli, Nicolai Karcher, Federica Armanini, Paolo Manghi, Kevin Bonham, Moreno Zolfo, … Nicola Segata
Cell Host & Microbe (2019-11) https://doi.org/ggc9dc
DOI: 10.1016/j.chom.2019.08.018 · PMID: 31607556 · PMCID: PMC6854460

17. CoproID predicts the source of coprolites and paleofeces using microbiome composition and host DNA content
Maxime Borry, Bryan Cordova, Angela Perri, Marsha Wibowo, Tanvi Prasad Honap, Jada Ko, Jie Yu, Kate Britton, Linus Girdland-Flink, Robert C. Power, … Christina Warinner
PeerJ (2020-04-17) https://doi.org/dr8x
DOI: 10.7717/peerj.9001 · PMID: 32337106 · PMCID: PMC7169968

18. Pathogens and host immunity in the ancient human oral cavity
Christina Warinner, João F Matias Rodrigues, Rounak Vyas, Christian Trachsel, Natallia Shved, Jonas Grossmann, Anita Radini, Y Hancock, Raul Y Tito, Sarah Fiddyment, … Enrico Cappellini
Nature Genetics (2014-02-23) https://doi.org/r4n
DOI: 10.1038/ng.2906 · PMID: 24562188 · PMCID: PMC3969750

19. Neanderthal behaviour, diet, and disease inferred from ancient DNA in dental calculus
Laura S. Weyrich, Sebastian Duchene, Julien Soubrier, Luis Arriola, Bastien Llamas, James Breen, Alan G. Morris, Kurt W. Alt, David Caramelli, Veit Dresely, … Alan Cooper
Nature (2017-03-08) https://doi.org/f9szrm
DOI: 10.1038/nature21674 · PMID: 28273061

20. Fifty thousand years of Arctic vegetation and megafaunal diet
Eske Willerslev, John Davison, Mari Moora, Martin Zobel, Eric Coissac, Mary E. Edwards, Eline D. Lorenzen, Mette Vestergård, Galina Gussarova, James Haile, … Pierre Taberlet
Nature (2014-02-05) https://doi.org/f2zr4s
DOI: 10.1038/nature12921 · PMID: 24499916

21. Neandertal and Denisovan DNA from Pleistocene sediments
Viviane Slon, Charlotte Hopfe, Clemens L. Weiß, Fabrizio Mafessoni, Marco de la Rasilla, Carles Lalueza-Fox, Antonio Rosas, Marie Soressi, Monika V. Knul, Rebecca Miller, … Matthias Meyer
Science (2017-05-12) https://doi.org/b6jd
DOI: 10.1126/science.aam9695 · PMID: 28450384

22. Plasmodium vivax Malaria Viewed through the Lens of an Eradicated European Strain
Lucy van Dorp, Pere Gelabert, Adrien Rieux, Marc de Manuel, Toni de-Dios, Shyam Gopalakrishnan, Christian Carøe, Marcela Sandoval-Velasco, Rosa Fregel, Iñigo Olalde, … Carles Lalueza-Fox
Molecular Biology and Evolution (2020-03) https://doi.org/ggqzq2
DOI: 10.1093/molbev/msz264 · PMID: 31697387 · PMCID: PMC7038659

23. Paging through history: parchment as a reservoir of ancient DNA for next generation sequencing
M. D. Teasdale, N. L. van Doorn, S. Fiddyment, C. C. Webb, T. O’Connor, M. Hofreiter, M. J. Collins, D. G. Bradley
Philosophical Transactions of the Royal Society B: Biological Sciences (2015-01-19) https://doi.org/ggqzq3
DOI: 10.1098/rstb.2013.0379 · PMID: 25487331 · PMCID: PMC4275887

24. A 5700 year-old human genome and oral microbiome from chewed birch pitch
Theis Z. T. Jensen, Jonas Niemann, Katrine Højholt Iversen, Anna K. Fotakis, Shyam Gopalakrishnan, Åshild J. Vågene, Mikkel Winther Pedersen, Mikkel-Holger S. Sinding, Martin R. Ellegaard, Morten E. Allentoft, … Hannes Schroeder
Nature Communications (2019-12-17) https://doi.org/ggfm6x
DOI: 10.1038/s41467-019-13549-9 · PMID: 31848342 · PMCID: PMC6917805

25. Ancient DNA from mastics solidifies connection between material culture and genetics of mesolithic hunter–gatherers in Scandinavia
Natalija Kashuba, Emrah Kırdök, Hege Damlien, Mikael A. Manninen, Bengt Nordqvist, Per Persson, Anders Götherström
Communications Biology (2019-05-15) https://doi.org/ggqzqz
DOI: 10.1038/s42003-019-0399-1 · PMID: 31123709 · PMCID: PMC6520363

26. The Beaker phenomenon and the genomic transformation of northwest Europe
Iñigo Olalde, Selina Brace, Morten E. Allentoft, Ian Armit, Kristian Kristiansen, Thomas Booth, Nadin Rohland, Swapan Mallick, Anna Szécsényi-Nagy, Alissa Mittnik, … David Reich
Nature (2018-02-21) https://doi.org/gcx74m
DOI: 10.1038/nature25738 · PMID: 29466337 · PMCID: PMC5973796

27. The genomic history of southeastern Europe
Iain Mathieson, Songül Alpaslan-Roodenberg, Cosimo Posth, Anna Szécsényi-Nagy, Nadin Rohland, Swapan Mallick, Iñigo Olalde, Nasreen Broomandkhoshbacht, Francesca Candilio, Olivia Cheronet, … David Reich
Nature (2018-02-21) https://doi.org/gc2n9h
DOI: 10.1038/nature25778 · PMID: 29466330 · PMCID: PMC6091220

28. A draft genome of Yersinia pestis from victims of the Black Death
Kirsten I. Bos, Verena J. Schuenemann, G. Brian Golding, Hernán A. Burbano, Nicholas Waglechner, Brian K. Coombes, Joseph B. McPhee, Sharon N. DeWitte, Matthias Meyer, Sarah Schmedes, … Johannes Krause
Nature (2011-10-12) https://doi.org/fk87wk
DOI: 10.1038/nature10549 · PMID: 21993626 · PMCID: PMC3690193

29. Bioinformatics Education–Perspectives and Challenges out of Africa
O. Tastan Bishop, E. F. Adebiyi, A. M. Alzohairy, D. Everett, K. Ghedira, A. Ghouila, J. Kumuthini, N. J. Mulder, S. Panji, H.-G. Patterton, (for the H3ABioNet Consortium, as members of The H3Africa Consortium)
Briefings in Bioinformatics (2014-07-02) https://doi.org/f67hjx
DOI: 10.1093/bib/bbu022 · PMID: 24990350 · PMCID: PMC4364068

30. Highlights on the Application of Genomics and Bioinformatics in the Fight Against Infectious Diseases: Challenges and Opportunities in Africa
Saikou Y. Bah, Collins Misita Morang’a, Jonas A. Kengne-Ouafo, Lucas Amenga–Etego, Gordon A. Awandare
Frontiers in Genetics (2018-11-27) https://doi.org/gfrxbz
DOI: 10.3389/fgene.2018.00575 · PMID: 30538723 · PMCID: PMC6277583

31. Instability and decay of the primary structure of DNA
Tomas Lindahl
Nature (1993-04) https://doi.org/d9c9vq
DOI: 10.1038/362709a0 · PMID: 8469282

32. Nuclear DNA sequences from the Middle Pleistocene Sima de los Huesos hominins
Matthias Meyer, Juan-Luis Arsuaga, Cesare de Filippo, Sarah Nagel, Ayinuer Aximu-Petri, Birgit Nickel, Ignacio Martínez, Ana Gracia, José María Bermúdez de Castro, Eudald Carbonell, … Svante Pääbo
Nature (2016-03-14) https://doi.org/bdcn
DOI: 10.1038/nature17405 · PMID: 26976447

33. Patterns of damage in genomic DNA sequences from a Neandertal
A. W. Briggs, U. Stenzel, P. L. F. Johnson, R. E. Green, J. Kelso, K. Prufer, M. Meyer, J. Krause, M. T. Ronan, M. Lachmann, S. Paabo
Proceedings of the National Academy of Sciences (2007-08-21) https://doi.org/bs4w7h
DOI: 10.1073/pnas.0704665104 · PMID: 17715061 · PMCID: PMC1976210

34. A new model for ancient DNA decay based on paleogenomic meta-analysis
Logan Kistler, Roselyn Ware, Oliver Smith, Matthew Collins, Robin G. Allaby
Nucleic Acids Research (2017-06-20) https://doi.org/gf58ts
DOI: 10.1093/nar/gkx361 · PMID: 28486705 · PMCID: PMC5499742

35. A Robust Framework for Microbial Archaeology
Christina Warinner, Alexander Herbig, Allison Mann, James A. Fellows Yates, Clemens L. Weiß, Hernán A. Burbano, Ludovic Orlando, Johannes Krause
Annual Review of Genomics and Human Genetics (2017-08-31) https://doi.org/gf5wqv
DOI: 10.1146/annurev-genom-091416-035526 · PMID: 28460196 · PMCID: PMC5581243

36. mapDamage2.0: fast approximate Bayesian estimates of ancient DNA damage parameters
Hákon Jónsson, Aurélien Ginolhac, Mikkel Schubert, Philip L. F. Johnson, Ludovic Orlando
Bioinformatics (2013-07) https://doi.org/gb5g2t
DOI: 10.1093/bioinformatics/btt193 · PMID: 23613487 · PMCID: PMC3694634

37. Early Divergent Strains of Yersinia pestis in Eurasia 5,000 Years Ago
Simon Rasmussen, Morten Erik Allentoft, Kasper Nielsen, Ludovic Orlando, Martin Sikora, Karl-Göran Sjögren, Anders Gorm Pedersen, Mikkel Schubert, Alex Van Dam, Christian Moliin Outzen Kapel, … Eske Willerslev
Cell (2015-10) https://doi.org/f3mxqd
DOI: 10.1016/j.cell.2015.10.009 · PMID: 26496604 · PMCID: PMC4644222

38. Characterization of ancient and modern genomes by SNP detection and phylogenomic and metagenomic analysis using PALEOMIX
Mikkel Schubert, Luca Ermini, Clio Der Sarkissian, Hákon Jónsson, Aurélien Ginolhac, Robert Schaefer, Michael D Martin, Ruth Fernández, Martin Kircher, Molly McCue, … Ludovic Orlando
Nature Protocols (2014-04-10) https://doi.org/f5x3qm
DOI: 10.1038/nprot.2014.063 · PMID: 24722405

39. EAGER: efficient ancient genome reconstruction
Alexander Peltzer, Günter Jäger, Alexander Herbig, Alexander Seitz, Christian Kniep, Johannes Krause, Kay Nieselt
Genome Biology (2016-03-31) https://doi.org/ggqzpk
DOI: 10.1186/s13059-016-0918-z · PMID: 27036623 · PMCID: PMC4815194

40. Fast and accurate short read alignment with Burrows-Wheeler transform
H. Li, R. Durbin
Bioinformatics (2009-05-18) https://doi.org/dqt59j
DOI: 10.1093/bioinformatics/btp324 · PMID: 19451168 · PMCID: PMC2705234

41. mapDamage: testing for damage patterns in ancient DNA sequences
Aurelien Ginolhac, Morten Rasmussen, M. Thomas P. Gilbert, Eske Willerslev, Ludovic Orlando
Bioinformatics (2011-08-01) https://doi.org/cn45v7
DOI: 10.1093/bioinformatics/btr347 · PMID: 21659319

42. The Stone Age Plague and Its Persistence in Eurasia
Aida Andrades Valtueña, Alissa Mittnik, Felix M. Key, Wolfgang Haak, Raili Allmäe, Andrej Belinskij, Mantas Daubaras, Michal Feldman, Rimantas Jankauskas, Ivor Janković, … Johannes Krause
Current Biology (2017-12) https://doi.org/cgmv
DOI: 10.1016/j.cub.2017.10.025 · PMID: 29174893

43. Novel Substrates as Sources of Ancient DNA: Prospects and Hurdles
Eleanor Green, Camilla Speller
Genes (2017-07-13) https://doi.org/gf57tz
DOI: 10.3390/genes8070180 · PMID: 28703741 · PMCID: PMC5541313

44. Nextflow enables reproducible computational workflows
Paolo Di Tommaso, Maria Chatzou, Evan W Floden, Pablo Prieto Barja, Emilio Palumbo, Cedric Notredame
Nature Biotechnology (2017-04-11) https://doi.org/gfj52z
DOI: 10.1038/nbt.3820 · PMID: 28398311

45. The nf-core framework for community-curated bioinformatics pipelines
Philip A. Ewels, Alexander Peltzer, Sven Fillinger, Harshil Patel, Johannes Alneberg, Andreas Wilm, Maxime Ulysse Garcia, Paolo Di Tommaso, Sven Nahnsen
Nature Biotechnology (2020-02-13) https://doi.org/ggk3qh
DOI: 10.1038/s41587-020-0439-x · PMID: 32055031

46. Conda — Conda documentationhttps://docs.conda.io/en/latest/

47. Empowering App Development for Developers | Dockerhttps://www.docker.com/

48. Home
Sylabs.io
https://sylabs.io/

49. Nextflow Towerhttps://tower.nf/

50. Improving ancient DNA read mapping against modern reference genomes
Mikkel Schubert, Aurelien Ginolhac, Stinus Lindgreen, John F Thompson, Khaled AS AL-Rasheid, Eske Willerslev, Anders Krogh, Ludovic Orlando
BMC Genomics (2012) https://doi.org/gb3ff7
DOI: 10.1186/1471-2164-13-178 · PMID: 22574660 · PMCID: PMC3468387

51. Babraham Bioinformatics - FastQC A Quality Control tool for High Throughput Sequence Datahttps://www.bioinformatics.babraham.ac.uk/projects/fastqc/

52. AdapterRemoval v2: rapid adapter trimming, identification, and read merging
Mikkel Schubert, Stinus Lindgreen, Ludovic Orlando
BMC Research Notes (2016-02-12) https://doi.org/gfzqhb
DOI: 10.1186/s13104-016-1900-2 · PMID: 26868221 · PMCID: PMC4751634

53. Fast and accurate long-read alignment with Burrows–Wheeler transform
Heng Li, Richard Durbin
Bioinformatics (2010-03-01) https://doi.org/cm27kg
DOI: 10.1093/bioinformatics/btp698 · PMID: 20080505 · PMCID: PMC2828108

54. Aligning sequence reads, clone sequences and assembly contigs with BWA-MEM
Heng Li
arXiv (2013-05-28) https://arxiv.org/abs/1303.3997

55. The Sequence Alignment/Map format and SAMtools
H. Li, B. Handsaker, A. Wysoker, T. Fennell, J. Ruan, N. Homer, G. Marth, G. Abecasis, R. Durbin, 1000 Genome Project Data Processing Subgroup
Bioinformatics (2009-06-08) https://doi.org/ff6426
DOI: 10.1093/bioinformatics/btp352 · PMID: 19505943 · PMCID: PMC2723002

56. The Genome Analysis Toolkit: A MapReduce framework for analyzing next-generation DNA sequencing data
A. McKenna, M. Hanna, E. Banks, A. Sivachenko, K. Cibulskis, A. Kernytsky, K. Garimella, D. Altshuler, S. Gabriel, M. Daly, M. A. DePristo
Genome Research (2010-07-19) https://doi.org/bnzbn6
DOI: 10.1101/gr.107524.110 · PMID: 20644199 · PMCID: PMC2928508

57. Predicting the molecular complexity of sequencing libraries
Timothy Daley, Andrew D Smith
Nature Methods (2013-02-24) https://doi.org/gfx6f5
DOI: 10.1038/nmeth.2375 · PMID: 23435259 · PMCID: PMC3612374

58. Qualimap 2: advanced multi-sample quality control for high-throughput sequencing data
Konstantin Okonechnikov, Ana Conesa, Fernando García-Alcalde
Bioinformatics (2015-10-01) https://doi.org/ggxrmx
DOI: 10.1093/bioinformatics/btv566 · PMID: 26428292 · PMCID: PMC4708105

59. Integrative-Transcriptomics/DamageProfiler: DamageProfiler v0.4.9
Judith Neukamm, Alexander Peltzer
Zenodo (2019-11-29) https://doi.org/ggxrmz
DOI: 10.5281/zenodo.3557708

60. Separating endogenous ancient DNA from modern day contamination in a Siberian Neandertal
Pontus Skoglund, Bernd H. Northoff, Michael V. Shunkov, Anatoli P. Derevianko, Svante Pääbo, Johannes Krause, Mattias Jakobsson
Proceedings of the National Academy of Sciences (2014-02-11) https://doi.org/f2z5sw
DOI: 10.1073/pnas.1318934111 · PMID: 24469802 · PMCID: PMC3926038

61. An efficient and scalable analysis framework for variant extraction and refinement from population-scale DNA sequence data
Goo Jun, Mary Kate Wing, Gonçalo R. Abecasis, Hyun Min Kang
Genome Research (2015-06) https://doi.org/f7dz2d
DOI: 10.1101/gr.176552.114 · PMID: 25883319 · PMCID: PMC4448687

62. Bioconda: sustainable and comprehensive software distribution for the life sciences
Björn Grüning, Ryan Dale, Andreas Sjödin, Brad A. Chapman, Jillian Rowe, Christopher H. Tomkins-Tinch, Renan Valieris, Johannes Köster, The Bioconda Team
Nature Methods (2018-07-02) https://doi.org/gd2xzp
DOI: 10.1038/s41592-018-0046-7 · PMID: 29967506

63. conda-forge | community driven packaging for condahttps://conda-forge.org/

64. Schmutzi: estimation of contamination and endogenous mitochondrial consensus calling for ancient DNA
Gabriel Renaud, Viviane Slon, Ana T. Duggan, Janet Kelso
Genome Biology (2015-10-12) https://doi.org/f72mvg
DOI: 10.1186/s13059-015-0776-0 · PMID: 26458810 · PMCID: PMC4601135

65. ANGSD: Analysis of Next Generation Sequencing Data
Thorfinn Sand Korneliussen, Anders Albrechtsen, Rasmus Nielsen
BMC Bioinformatics (2014-11-25) https://doi.org/gb8wpz
DOI: 10.1186/s12859-014-0356-4 · PMID: 25420514 · PMCID: PMC4248462

66. Fast gapped-read alignment with Bowtie 2
Ben Langmead, Steven L Salzberg
Nature Methods (2012-03-04) https://doi.org/gd2xzn
DOI: 10.1038/nmeth.1923 · PMID: 22388286 · PMCID: PMC3322381

67. Assessing DNA Sequence Alignment Methods for Characterizing Ancient Genomes and Methylomes
Marine Poullet, Ludovic Orlando
Frontiers in Ecology and Evolution (2020-05-06) https://doi.org/ggzwqr
DOI: 10.3389/fevo.2020.00105

68. fastp: an ultra-fast all-in-one FASTQ preprocessor
Shifu Chen, Yanqing Zhou, Yaru Chen, Jia Gu
Bioinformatics (2018-09-01) https://doi.org/gd9mrb
DOI: 10.1093/bioinformatics/bty560 · PMID: 30423086 · PMCID: PMC6129281

69. QC Fail Sequencing » Illumina 2 colour chemistry can overcall high confidence G baseshttps://sequencing.qcfail.com/articles/illumina-2-colour-chemistry-can-overcall-high-confidence-g-bases/

70. Haplotype-based variant detection from short-read sequencing
Erik Garrison, Gabor Marth
arXiv (2012-07-24) https://arxiv.org/abs/1207.3907

71. MALT: Fast alignment and analysis of metagenomic DNA sequence data applied to the Tyrolean Iceman
Alexander Herbig, Frank Maixner, Kirsten I. Bos, Albert Zink, Johannes Krause, Daniel H. Huson
bioRxiv (2016-04-27) https://doi.org/ggxkk9
DOI: 10.1101/050559

72. Salmonella enterica genomes from victims of a major sixteenth-century epidemic in Mexico
Åshild J. Vågene, Alexander Herbig, Michael G. Campana, Nelly M. Robles García, Christina Warinner, Susanna Sabin, Maria A. Spyrou, Aida Andrades Valtueña, Daniel Huson, Noreen Tuross, … Johannes Krause
Nature Ecology & Evolution (2018-01-15) https://doi.org/ggxkk7
DOI: 10.1038/s41559-017-0446-6 · PMID: 29335577

73. Improved metagenomic analysis with Kraken 2
Derrick E. Wood, Jennifer Lu, Ben Langmead
Genome Biology (2019-11-28) https://doi.org/ggfk55
DOI: 10.1186/s13059-019-1891-0 · PMID: 31779668 · PMCID: PMC6883579

74. HOPS: automated detection and authentication of pathogen DNA in archaeological remains
Ron Hübler, Felix M. Key, Christina Warinner, Kirsten I. Bos, Johannes Krause, Alexander Herbig
Genome Biology (2019-12-16) https://doi.org/ggxkmb
DOI: 10.1186/s13059-019-1903-0 · PMID: 31842945 · PMCID: PMC6913047

75. Microbial differences between dental plaque and historic dental calculus are related to oral biofilm maturation stage
Irina M. Velsko, James A. Fellows Yates, Franziska Aron, Richard W. Hagan, Laurent A. F. Frantz, Louise Loe, Juan Bautista Rodriguez Martinez, Eros Chaves, Chris Gosden, Greger Larson, Christina Warinner
Microbiome (2019-07-06) https://doi.org/ggxkmc
DOI: 10.1186/s40168-019-0717-3 · PMID: 31279340 · PMCID: PMC6612086

76. BEDTools: a flexible suite of utilities for comparing genomic features
Aaron R. Quinlan, Ira M. Hall
Bioinformatics (2010-03-15) https://doi.org/cmrms3
DOI: 10.1093/bioinformatics/btq033 · PMID: 20110278 · PMCID: PMC2832824

77. Ancient Fennoscandian genomes reveal origin and spread of Siberian ancestry in Europe
Thiseas C. Lamnidis, Kerttu Majander, Choongwon Jeong, Elina Salmela, Anna Wessman, Vyacheslav Moiseyev, Valery Khartanovich, Oleg Balanovsky, Matthias Ongyerth, Antje Weihmann, … Stephan Schiffels
Nature Communications (2018-11-27) https://doi.org/ggxkk6
DOI: 10.1038/s41467-018-07483-5 · PMID: 30479341 · PMCID: PMC6258758

78. MultiQC: summarize analysis results for multiple tools and samples in a single report
Philip Ewels, Måns Magnusson, Sverker Lundin, Max Käller
Bioinformatics (2016-10-01) https://doi.org/f3s996
DOI: 10.1093/bioinformatics/btw354 · PMID: 27312411 · PMCID: PMC5039924

79. eager » nf-corehttps://nf-co.re/eager/docs

80. nf-core/eager
nf-core
(2020-06-09) https://github.com/nf-core/eager

81. nf-core/configs
nf-core
(2020-06-10) https://github.com/nf-core/configs

82. Open collaborative writing with Manubot
Daniel S. Himmelstein, Vincent Rubinetti, David R. Slochower, Dongbo Hu, Venkat S. Malladi, Casey S. Greene, Anthony Gitter
PLOS Computational Biology (2019-06-24) https://doi.org/c7np
DOI: 10.1371/journal.pcbi.1007128 · PMID: 31233491 · PMCID: PMC6611653