Annotated Bibliography Using Refworks Rutgers

Abstract

The vast majority of microorganisms on Earth reside in often-inseparable environment-specific communities—microbiomes. Meta-genomic/-transcriptomic sequencing could reveal the otherwise inaccessible functionality of microbiomes. However, existing analytical approaches focus on attributing sequencing reads to known genes/genomes, often failing to make maximal use of available data. We created faser (functional annotation of sequencing reads), an algorithm that is optimized to map reads to molecular functions encoded by the read-correspondent genes. The mi-faser microbiome analysis pipeline, combining faser with our manually curated reference database of protein functions, accurately annotates microbiome molecular functionality. mi-faser’s minutes-per-microbiome processing speed is significantly faster than that of other methods, allowing for large scale comparisons. Microbiome function vectors can be compared between different conditions to highlight environment-specific and/or time-dependent changes in functionality. Here, we identified previously unseen oil degradation-specific functions in BP oil-spill data, as well as functional signatures of individual-specific gut microbiome responses to a dietary intervention in children with Prader–Willi syndrome. Our method also revealed variability in Crohn's Disease patient microbiomes and clearly distinguished them from those of related healthy individuals. Our analysis highlighted the microbiome role in CD pathogenicity, demonstrating enrichment of patient microbiomes in functions that promote inflammation and that help bacteria survive it.

INTRODUCTION

Microorganisms inhabit every available niche of our planet, and our bodies are no exception. Microbes that survive and thrive in the environments at the extremes of temperature, pH, and chemical or radiation contamination possess unique molecular functions of high industrial, clinical, and bioremediation value. The human body microbiome critically impacts our health. For example, Crohn's disease (CD) is a multifactorial disorder resulting from the interplay of individual genetic susceptibility, the gastrointestinal (GI) microbiome and other environmental factors. Taxonomic surveys of the GI microbiome have revealed microbial community features that are unique to CD patients, e.g. overall loss of microbial diversity (1,2), as well as depletion and enrichment of certain bacterial taxa (3–6). Establishing whether these observed microbial community shifts contribute to pathogenesis or, instead, correlate with or result from the disease onset, requires understanding not only what are the microbes involved, but also what they do. Earlier studies indicate that in association with CD, the microbiome molecular function potential is more consistently disturbed than taxonomic makeup (7). More thorough functional analyses, e.g. based on deep metagenomic sequencing, are necessary to elucidate these findings.

Metagenome functional annotation can be performed with or without genome assembly. If the reads can be assembled into large contigs, existing annotation pipelines, such as RAST (8) and IMG (9), can be applied. However, assembly is difficult and often plagued by a large fraction of unassembled reads or short length contigs, which belong to the minor microbiome members, and by chimeric assemblies, which are especially common for complex and highly diverse samples (see Sczyrba et al., 2017, doi: https://doi.org/10.1101/099127). Downstream gene finding algorithms are further faced with incomplete and erroneously assembled sequences, complicating statistical model constructions. Read-based annotation, e.g. using a platform such as MG-RAST (10), can access molecular functionality of the entire community. However, reads are usually annotated via function transfer by homology that, due to the short read length, is lacking in precision. This inaccuracy is additionally compounded by the erroneous computational annotations of most genes in the reference databases (11).

Here, we compiled a gold standard set of reference proteins (GS), with experimentally annotated molecular functions. We further developed faser (functional annotation of sequencing reads), an algorithm that uses alignments of translated sequencing reads to full-length proteins to annotate read-‘parent protein’ molecular functionality. faser annotates reads with higher precision at higher resolution, i.e. more specific functionality, than BLAST or PSI-BLAST. In a benchmark test, the functional annotations produced by the combination of the faser algorithm with the GS database were 12% more accurate than MG-RAST. Note that this performance may be an overestimate because the benchmark metagenome included the GS database. However, when GS was replaced with md5nr, MG-RAST’s reference database, faser annotated 20% more reads than MG-RAST at a comparable precision level. These results illustrate that the GS and faser combination improves on MG-RAST capabilities.

Our mi-faser pipeline implementation (Figure 1), combining faser and GS, is highly parallelized, making use of all available compute cores and processing a (∼10GB/70M read) meta-genomic/-transcriptomic file in under half an hour (using 400 compute cores, on average). Note that if multiple microbiomes are submitted for annotation in parallel, the time scales favourably; in testing, 17 metagenomes were processed within 66 minutes. mi-faser results for all microbiomes analysed in this manuscript are available at http://services.bromberglab.org/mifaser/results/example. The standalone version of the pipeline, along with the mi-faser source code, is available at https://bitbucket.org/bromberglab/mifaser, as well as on the bromberglab website.

Figure 1.

mi-faser pipeline. mi-faser is parallelized and runs a load balancer to submit jobs to available [1–2000] compute cores. Under normal functioning conditions (∼400 available cores, on average), it takes ∼30 min to process a single (10G/70M read) meta-genome/-transcriptome.

Figure 1.

mi-faser pipeline. mi-faser is parallelized and runs a load balancer to submit jobs to available [1–2000] compute cores. Under normal functioning conditions (∼400 available cores, on average), it takes ∼30 min to process a single (10G/70M read) meta-genome/-transcriptome.

We applied our mi-faser to metagenomic data collected from beach sands in different stages of oil contamination (12). Here, mi-faser was able to identify oil degradation functionality that was missed by MG-RAST. We further performed large-scale analysis of 68 metagenomic datasets from a study of dietary intervention in Prader-Willi syndrome (PWS) affected obese children. Each dataset was processed in approximately 16 minutes, highlighting mi-faser

Rutgers off-campus users must log-in for access
to Rutgers restricted resources linked from RefWorks.



Introducing the new RefWorks.
Still using the old version of RefWorks? Upgrade to the newest version to take advantage of the latest features. For more information, see Upgrading from Legacy RefWorks to the New RefWorks.


What is RefWorks 3?

RefWorks 3 is the latest version of RefWorks, a research management tool that makes it easy to save references to books, articles, and web pages and organize them into your own personal research library. With RefWorks 3 you can:

  • Save references and full text documents from major online databases and websites
  • Access your references and documents from any computer or mobile device
  • Organize, highlight, and annotate your documents
  • Collaborate and share document collections with other RefWorks users
  • Add citations and footnotes to your research paper from references in your RefWorks library
  • Automatically create a bibliography or works cited in any citation style

How do I create an account?

To create an account:

  1. Go to and click “Create Account”
  2. Enter your Rutgers email address and create a password
  3. Click the activation link in the confirmation email to complete your account setup

How is RefWorks 3 different from legacy RefWorks?

RefWorks 3 offers a variety of new and improved features such as easy one-click importing, automatic recognition of uploaded documents, and document sharing, highlighting, and annotation. For more information, visit the RefWorks Support Center.


Will RefWorks 3 replace the legacy RefWorks?

Eventually. The legacy RefWorks platform is expected to be retired in 2018. Legacy RefWorks users should plan to upgrade to the new platform before then.


How do I upgrade to the newest version of RefWorks?

To upgrade from legacy RefWorks:

  1. Login to RefWorks
  2. Click the link at the top of the screen that says “Move to the newest version of RefWorks”
  3. Follow the on-screen prompts to compete your account setup

Note: Upgrading from the legacy RefWorks is not recommended if you are in the middle of a major project. Write-n-Cite papers started in legacy RefWorks may not be compatible with new RefWorks. It is advisable that you wait to upgrade until after completing your work. For more information, see the Upgrading from Legacy RefWorks to the New RefWorks.


Can I use both versions of RefWorks simultaneously?

Yes, legacy RefWorks will continue to be available until 2018. This will allow you to choose a time to migrate that works best for you. In the meantime, you can try out RefWorks 3 by creating a new RefWorks account and migrating your old RefWorks account later.

A few notes about using RefWorks and RefWorks 3 simultaneously:

  • Your RefWorks libraries will not automatically be synced. However, you can migrate your legacy RefWorks library at any time. See How do I upgrade to the newest version of RefWorks?
  • Papers written with previous versions of Write-n-Cite will not be compatible with the new version of RefWorks.

How do I save references to RefWorks?

RefWorks has an easy to use tool that installs in any web browser called “Save to RefWorks”. This browser bookmarklet grabs information from any web page – in some cases a complete article – and adds it to your RefWorks library.

To add the “Save to RefWorks” bookmarklet to your browser:

  1. Click on your name in the upper right-hand corner and select “Tools”
  2. Click “Install Save to RefWorks”
  3. Drag the “Save to RefWorks” button to your browser’s Favorites bar
  4. Go to any web page and click “Save to RefWorks” to capture information

“Save to RefWorks” is not supported by all sites. Results vary by site. For more information, see Using Save to RefWorks to capture citations, articles and full-text on the web.

For databases that do not support the “Save to RefWorks” bookmarklet, you can still save references by using the “Export to RefWorks” function within the database. For more information, see Using Export to RefWorks to import citations into your library.

You can also add items to your RefWorks library by uploading documents from your computer. RefWorks uses its document recognition feature to attempt to auto-complete the citation information. For more information, see Uploading one or more documents from your computer.

References can also be added or edited manually. For more information, see Adding references manually.


How do I save references from the library catalog to RefWorks?

To save references from the Rutgers library catalog:

  1. Select one or more references from your search results and add them to your Book Bag
  2. Click Book Bag to view your selected items
  3. Select the items you wish to export and click “Export Selected”
  4. Choose RefWorks as your format option and click “Export”
  5. Login to your RefWorks account to complete the export

You can also save individual references using the “Save to RefWorks” bookmarklet, but the direct export option works much better.


How do I import references from other citation managers?

To import references from other citation managers into RefWorks, follow the instructions below:


How do I create a bibliography?

To create a bibliography:

  1. Select the collection or group of references you want to use
  2. Click the “Bibliography” icon and select “Create Bibliography”
  3. Choose your citation style (APA, MLA, Chicago, etc.)
  4. Click “Copy to Clipboard” and paste your bibliography into your document

How do I insert citations and footnotes into my research paper?

RefWorks offers plugins for Microsoft Word and Google Docs that allow you to insert citations and footnotes from references in your RefWorks library. To install these add-ons, click on your name in the upper right-hand corner and select “Tools”. For instructions on using these tools, see:

Check compatibility requirements before installing Write-N-Cite. For installation and troubleshooting assistance, contact RefWorks Technical Support at refworks.support@proquest.com.


Can I create or modify citation styles?


How do I create a collection?

Create collections or organize items in your RefWorks library. Items can be assigned to multiple collections, but they only need to be added to RefWorks once.

To create a collection:

  1. Click “My Collections” and select “Add a Collection”
  2. Name your collection and click “Save”
  3. [Optional] Click the More icon next to your collection to add subcollections
  4. Drag and drop items to add them to your new collection

How do I share collections with other RefWorks users?

RefWorks allows you to share collections with other RefWorks users at Rutgers or outside of Rutgers (as long as the other person has an institutional RefWorks account).

To share a collection with other RefWorks users:

  1. Select the collection you want to share
  2. Click the “Share & Export” icon and select “Share Collection”
  3. Enter the email address of the person you want to share with (one at time)
  4. Select the level of access for that person (read, annotate, or modify)
  5. Click “Share Collection”

The person(s) you've invited will receive an email notification (and a notification inside their RefWorks account) where they can accept the invitation. For more information, see Sharing a collection with other RefWorks users.


How do I annotate documents?

RefWorks has an integrated document reader that allows you read, highlight, and annotate full text documents.

To annotate a document:

  1. Select the reference with the document and click “Read”
  2. To highlight text or add a comment, click on the appropriate icon from the top menu and select an area of text

Comments will appear to the far right of the selected area. You can change the color of your comments by clicking the More option for the comment and selecting a new color. You can also edit or delete the comment. Once you type a note, it is automatically saved. For more information, see Annotating documents or articles.


Does RefWorks have a storage limit?

RefWorks allows you to store up to 100GB worth of documents. You can also choose to store your documents in Dropbox. For more information, see RefWorks integration with Dropbox.


Will I continue to have access to my RefWorks account after I leave Rutgers?

Yes, all faculty and student alumni will continue to have access to their RefWorks account as long as Rutgers continues to subscribe.


Is RefWorks compatible with all web browsers?

RefWorks 3 works best on Chrome, Firefox, and Safari. Some features may not work with Internet Explorer. For more information, see compatibility requirements.


Where can I find more information about RefWorks?

0 thoughts on “Annotated Bibliography Using Refworks Rutgers”

    -->

Leave a Comment

Your email address will not be published. Required fields are marked *