AlgaeDB Documentation

Homepage

The AlgaeDB Homepage acts as a hub that allows users to access all other sections of the website. There are three main sections in the homepage:

1. The navigation bar (navbar).
2. Navbar search tool.
3. Navigation sidebar.
4. Assembly search table.

The navbar search tool allows users to search AlgaeDB for assemblies, pages, and genes.

The assembly search table displays all the assemblies contained in AlgaeDB in an interactive table, powered by DataTables, along with a brief summary of each assembly. For more information on how to use interactive tables, please view the interactive table help section. The table can be filtered using the search bar at the top right of the table. Each column can be ordered in either ascending or descending order. The final column of the table contains allows users to download available files for each assembly using a dropdown menu. Clicking on the button will show the available files, and clicking each one will download the file.

While 11 columns are shown, this table contains numerous hidden columns, which can be used to search the available assemblies. In addition to the visible columns, the following fields are also searchable in the assembly table:

• Phylum
• Subphylum
• Subclass
• Family
• Subfamily
• Order
• Suborder
• Tribe
• Publication name
• Publication authors
• Publication journal
• Species authority
• Genus authority
• Bioproject
• Biosample
• Sequencing technology
• Strain

The AlgaeDB homepage displays an interactive world map showing the locations of assemblies, powered by Plotly. Not all assemblies have location data available, and thus the map will not show all assemblies. The map orientation can be moved by clicking and dragging, while the mouse scroller can be used to zoom in or out. Plot controls can be seen on the top right while the cursor is in the plot area. By clicking on the Box Select or Lasso Select buttons, plot points can be selected, with those not selected being grayed out. Double clicking resets the selection.

Search

AlgaeDB contains two main search functions: a quick search function available at the top of the page through the navigation bar, and an advanced search available through the search page. The quick search function can be used to search for assemblies and pages, and can redirect the user automatically to relevant pages when inputting the name of a page or an assembly profile. For example, searching 'blast' will automatically redirect the user to the BLAST page, which searching for Chondrus crispus Stackhouse will redirect the user to that Annotations Explorer page. Otherwise, searching in the quick search bar will redirect the user to the quick search page, which will display results for both assemblies and pages.

An advanced search function is also available via the search page. This can be used to construct detailed queries to search the AlgaeDB database. To use this page, enter a search query into the search fields. Particular fields can be selected to be searched using the 'Select field' option, while advanced operators such as AND, OR, and NOT can be selected in the 'Select search operator' field. Additional rows can be added using the '+' button, and rows can be removed using the '-' button. This advanced search filter can be used to search either assemblies or individual genes. Fields for both searches are available with both types, so gene names can be used to search for assemblies, and vice versa.

BLAST

Input

Sequence similarity can be searched using the AlgaeDB BLAST interface. BLAST searches are run against the assembly databases stored on AlgaeDB, and can be refined further to BLAST against specific organisms. Results are then displayed on dynamic tables that can be searched for specific results and downloaded or copied to clipboard.

1. Input form. Paste nucleotide or protein sequences here. Sequences can be in either raw text or in FASTA format. When posting in FASTA format, multiple queries can be used. For example, a multiple fasta search would look like:
   >Seq1
   MPAEEQHYSLHGQLPGSVMSYVFQLAGDALTANPAVPLHRYVLAAQRLMREADRQGAMGNLLIRSVLLVRFVQL
   >Seq2
   MSSGYINPSGNSSGMNQATSSILSSDSQMEGVTSSDENSMMYQDHSQGNQQVNRYSIEHPQDNRLPAIDY
2. E-value threshold. The E-value (or Expect Value) describes how many hits a user could expect from a random search, and correlates with stringency. A smaller e-value described a closer hit than a larger e-value. A maximum BLAST e-value can be set using this field. E-values between 1e-100 and 10 are accepted. The default is 0.1.
3. Choose BLAST algorithm. Use this dropdown select field to choose a BLAST algorithm. The algorithms available include BLASTp (search protein against protein), BLASTn (search nucleotide against nucleotide), BLASTx (search nucleotide against protein) and tBLASTn (search protein against translated nucleotide sequences).
4. Insert Example FASTA. This button inserts an example protein FASTA into the searchfield.

Loading page

When ready, submit your BLAST search by clicking the Submit BLAST Search button. This will start the BLAST process. If successful, a job ID will be generated, and you will be redirected to a job loading page. The URL for this page will also be used to load your results, so keep a copy of this code if you wish to come back to your results later.

BLAST Results page

When your BLAST job is finished, you will be taken to the BLAST results page. This uses the same URL as the loading page, and can be viewed for up to a week after completion. The BLAST results page has six main sections.

1. Queries list. The names of BLAST queries are displayed here. If no FASTA name was given, the query will be displayed as "unnamed protein product" or "unnamed nucleotide product". Click on each query name to navigate to the query.
2. Query details. Details of the BLAST search are displayed here. These include:
   • Job ID.
   • BLAST algorithm.
   • Threshold value
   • Date and time the job was queued at.
   • Data and time the AlgaeDB began processing the BLAST job.
   • Data and time the BLAST job was completed.
3. Results download (.XML format). Click this button to download the raw XML results of the BLAST job.
4. Query results header. The results table for each query can be copied to the clipboard using the "Copy" button. Results can also be downloaded as an Excel table using the "Excel" button. The number of entries shown on each page of the results table can also be controlled here.
5. Results table. Results are displayed in table form. Each column can be ordered by clicking the diamonds at the top of each column. Hold shift while clicking to order by multiple columns. Columns include the gene description, subject name, length (base or amino acid), overall score, bit score, E-value, and percent identity.
6. Results search bar. Results can be searched or filtered by typing search terms in this bar.

Statistics

Functional annotation statistics can be explored using the various pages under the "Statistics" header. The following statistics pages can be viewed:

• BUSCO summaries
• BUSCO summaries
• Repeat element predictions

• Compare: browse sequence annotations from all assemblies
• Metabolic pathways: Compare completion status of different metabolic pathways between organisms

Annotations Exlorer

The annotations explorer allows users to explore the functional annotations of individual assemblies as well as search groups of assemblies. Individual annotations can be accessed through links on the Homepage or via the Search function. They can also be searched using the AlgaeDB ID code, or other identifying data, such as NCBI assembly accession codes, or assembly names. Groups of assemblies, or all assemblies, can be accessed through the Annotations Explorer: Search page.

Annotations Explorer: Individual assemblies

Annotation details of individual assemblies are viewed on dynamic pages. On the left, basic assembly details and statistics are shown, while on the right sample details, taken from NCBI Biosample are shown. These details are dynamic, and absent fields will not be shown. Above the Biosample table, a dynamic world map powered by Plotly shows the location of the assembly sample. If the assembly has no available location data, then the map will state "No Location Data Available". Below the Biosample details, a downloads button will be diplayed if downloads are available for the file. This activates a dropdown menu of available files. Clicking on each item in the dropdown list will begin the download of the file.

Annotations can be explored using three search bars. Assemblies without functional annotations will not have these search functions available. These three bars can be used to search annotations by name or description, Enzyme Code, Gene Ontology and InterPro annotations, or KEGG annotations, respectively. These search bars can be used to query the functional annotation data, and function in the same way as the search page. The data that is retrieved is the same between all three functions, but is displayed differently depending on the table searched.

The three bars all search and retrieve the same data, but display different results depending on the search tab used. Results are displayed on an interactive table powered by DataTables (for more information on dynamic tables, see the Dynamic tables section).

Downloads

Downloads can either be accessed through the Downloads page, or through individual assembly pages. Each assembly profile can have one or more files available for download.

Batch downloads for each type of file can be downloaded from the Downloads page. These include all available files for each type of download, including assemblies, protein or CDS annotations, BUSCO results, repeat annotations, and annotation tables.

Downloads for each individual assembly profile can also be accessed from each profiles Annotation Explorer page, or through a dynamic table shown on the downloads page. This table can be searched using the same fields as the home page and search page tables, including species name, synonyms, taxonomy, and assembly level.

Accordions

Several pages in AlgaeDB use accordions to neatly sort large amounts of data. Accordions are collapsible tabs that can be expanded or collapsed to show or hide data, and are implemented using Bootstrap version 5.3.

Closed accordions hide additional content that can be shown by clicking on the accordion. Closed accordions have an off-white colour, and have a downwards pointing arrow. When open, the accordion header will be coloured in, and the arrow will point up.

Figures and maps

Figures and maps in AlgaeDB are implemented with Plotly in order to visualise data interactively. All figures and graphs displayed with plotly have interactive features. These include a control panel, which can be displayed by hovering the cursor over the graph or map, as well as interactive selections and movements.

In Plotly figures, plots such as bar charts can be controlled either with direct cursor interactions, or using the control panel. By hovering the cursor over the bars on the chart, further information is displayed. For example, the BUSCO page has a stacked bar chart displaying the completeness values of different assemblies. By hovering over each different coloured section, further data can be viewed.

Maps displayed with Plotly have the same core controls as figures. However, they have the following additional features:

• Rotation: maps can be rotated by clicking and dragging the cursor across the map. This rotates the map.
• Zoom: The map can be zoomed in and out of using the mouse wheel in addition to the zoom buttons on the control panel. Zooming in may change the shape of the map from the globe shape to a rectangular shape.

Dynamic Tables

Dynamic tables have been implemented in AlgaeDB using the DataTables library. Dynamic tables have several interactive features. Not all features are availabe for all dynamic tables in AlgaeDB, and not all tables are dynamic.

Dynamic tables have several interactive features that allow users to search and interact with the data, though not all tables will have all features available. The available features for these tables are:

1. Data copy and download controls. These buttons allow the user to copy the table to the clipboard, or to download the table as an Excel or as a CSV file. Note that not all interactive tables have these buttons available.
2. Pagination controls. Use this dropdown menu to select how many items are displayed on each page of the table.
3. Search controls. Search the data in the table using this search bar. Note that some tables have hidden columns which are not visible, but can be searched.
4. Table column controls. Columns can be ordered by clicking on the up and down arrows above each column. The table can be ordered using multiple columns by holding shift and clicking on multiple control arrows.
5. Pagination information. This displays how many items are being displayed in the current page of the table and the total number of items.
6. Pagination controls. Traverse different pages of the table using the pagination controls.

Data Collection

Data used in this database is collected from publicly available sources including the NCBI Genome database (formerly the NCBI Assembly database), as well as project datasets, as detailed in the Datasets section. Assemblies with protein annotations available were downloaded from their respective sources and their protein annotations used as it. Assemblies without protein annotations were downloaded, and functionally annotated (described below).

Quality control

Assembly completeness is quantified using BUSCO (Benchmarking Universal Single Copy Orthologues). Genome assemblies are graded for quality in both genome mode, using the assembly, and in protein mode, which uses the protein annotations. Transcriptome assemblies were graded in transcriptome and in protein mode. In genome and protein mode, BUSCO was run using AUGUSTUS as the gene predictor, with the species set to Galdieria. The training data generated by BUSCO was then used to train AUGUSTUS for downstream gene prediction. For transcriptome assemblies, which cannot be run with AUGUSTUS, BUSCO was then run in genome mode an additional time in order to generate downstream training data.

Functional-annotation

Gene inference

Genes were inferred for assemblies using AUGUSTUS. Genome assemblies first had repetitive regions soft-masked with RepeatMasker using custom repeat libraries constructed using RepeatModeler. Transcriptome assemblies were not masked. Where assemblies already had gene annotations published, those gene annotations were used instead. However, repeat masking was still performed on those assemblies in order to provide a comparison.

Taxonomy

The taxonomic identities of assemblies in this database are taken from the publication source of the assembly and crossreferenced against the most current taxonomy listed in AlgaeBase. Where confusion or inconsistencies arise, further investigation is undertaken to determine the identity of the assembly. If no further information can be found, the identity listed by the source or latest release will be used, but information will be provided to alert the user to any taxonomic uncertainties.