You will receive a six character alphanumeric analysis code after
submission. We recommend saving this code before you close the app. You
will need to enter it later on the Retrieve Results tab to
view your results.
Before submitting you can provide an email to receive a link to your
analysis. The app will also save any analyses you’ve submitted on your
current device (laptop, phone, etc.), and will list them under the
Retrieve Results tab.
Query Data
Data Table
The Data Table tab shows the processed input data in
tabular form. The default view includes query name, the species and
lineage in which it is found, and Pfam domain architecture, but the
table can be customized with the Add/remove column(s)
button. Columns can be filtered for particular species, lineages,
percentages, etc., and the entire table can be searched with plain text
or regex. The full data table can be downloaded in .csv
format with the Download as csv button.
FASTA
All FASTA sequences for the query protein(s) are provided for ease of
access.
Query Heatmap
A heatmap shows the occurrence of query protein(s) by taxonomic
presence, which may be useful for multi-FASTA input of homologs.
Query Domain Architecture
A customizable domain architecture visualization shows the query
protein(s) grouped by analysis or query. You can modify the domain plot
by selecting the Analysis box and adding or removing
results to display (e.g., Pfam, Phobius, Coils analyses).
Homolog Data
The Homolog Data table lists the best hits from all superkingdoms of
life (queried across all refseq or nr
genomes). Like in Query Data > Data Table described
above, tabular details are provided across all homologs, including
genome, species, lineage, and domain architecture information. Many
homology-specific options available in Add/remove column(s)
like percent identity, cluster ID (BLAST parameters). The accession
number for each homolog is linked to its corresponding NCBI protein
page.
Domain Architecture
A protein’s domain architecture (DA) refers to the order of specific
functional regions of a protein. Currently, MolEvolvR uses
databases and prediction algorithms integrated with InterProScan to
characterize the domain architecture of protein queries and their
homologs. We summarize the data with a set of useful visualizations
below. Results from Pfam, Phobius, Gene3D, SignalP_Gram_positive,
SignalP_Gram_negative, MobiDBlite, Hamap, and Coils are available.
Table
The table provides summary statistics on the domain architecture data
across all homologs, with the top (most frequent) domain architectures
by query protein (or across all queries) and the frequencies of
occurrence and lineages in which they occur. Click each row to view the
domain architecture spread across lineages. A popup demonstrates the
‘LineageCount’ by showing the frequencies of occurrence by individual
lineage for the selected domain architecture.
Heatmap
A color gradient heatmap across the query protein(s) indicates the
number of homologs identified within each lineage per domain
architecture.
Rows: Predominant domain architectures.
Columns: Key lineages from across the superkingdoms of
life.
Network
A network visualization summarizes domain architectures across query
protein(s) and their homologs Nodes represent a domain and edges denote
domain co-occurrence within a protein. The domains (nodes) that co-occur
within a protein/domain architecture are connected (edges), and the size
of nodes and thickness of edges are proportional to their relative
occurrences across homologs (or query proteins).
Interproscan Visualization
Each column of this visualization is organized by the database the
domains were obtained from. The rows represent select query protein(s)
and/or homologs (if a homology search was performed) with the lineage
added to the front of the accession number. You can select specific
proteins with the dropdown box, and choose group rows by analysis or
query. The visualization can also be updated by toggling available
database options under Analysis, or by adjusting the
Total Cutoff Count slider.
UpSet plot
An UpSet plot is a helpful summary visualization that shows the
frequencies of domains and domain architectures across all homologs. It
shows distribution of constituent domains underlying all homologs in a
histogram (to the left). The combination matrix displays the various
combinations of domains present across the domain architectures. The
adjoining second histogram (on top) shows the frequency of occurrences
of the indicated domain architectures (combinations).
Lineage Bar Chart
This stacked bar chart shows the frequency of occurrences of the top
domain architectures per lineage; the domain architectures for each
lineage-wise splits are displayed on the left.
Phylogeny
Phylogenetic analysis of proteins provides key insights into their
development and evolution. The conservation of certain portions through
lineages or across domains of life could indicate the importance of the
protein in certain biological processes.
Sunburst
An interactive sunburst plot shows the phyletic spread of the query
protein (selected with the Protein dropdown) across life. Hovering over
each section of the plot displays the lineage. The depth of displayed
taxonomic levels can be adjusted with Number of Levels to
add more detail to the sunburst plot.
Tree
This visualization is constructed from a multiple sequence alignment
of representative homologs. Tree leaves are labeled by lineage, species
(three-letter abbreviation), and accession numbers.
You can adjust the tree generation in two important ways: based on
whether homologs are reduced to representative sequences (e.g., by
lineage, species, or domain architecture), and based on the multiple
sequence alignment (MSA) algorithm chosen (including Clustal Omega,
Clustal W, and Muscle). The size of the tree can be altered selecting
the desirable number of sequences. To the right of the tree is a
visualization of the multiple sequence alignment, colored by amino acid
and showing overall conservation of sequence and structure of the
homologs used in tree construction.
MSA
You can customize and download a multiple sequence alignment as a
.pdf file, including a user-specified number of
representative sequences among the homologs.
