ecg

Pulling information from biological databases, and converting it into easy to use jsons/gmls for network science.

ecg

Dependencies

docopt^k,j,e (for CLI)
tqdm^k,j (for visual progress bars)
biopython^k (for KEGG REST API and TogoWS)
selenium^j (for webdriver)
beautifulsoup4^j (for web page parsing)
networkx^e (for graphs/networks

^kUsed by kegg.py

^jUsed by jgi.py

^eUsed by ecg.py

Quickstart

Open your terminal or other Unix/Linux command-line interface. Use it to navigate to your desktop, documents, or other folder in which you tend to store projects (e.g. cd Desktop/). Then, copy+paste into the terminal each of the following lines:

mkdir ecgHub
pip install docopt; pip install tqdm; pip install biopython; pip install selenium; pip install beautifulsoup4; pip install networkx
cd ecgHub
git clone https://github.com/ELIFE-ASU/ecg
cd ecg
pip install -e .
mkdir mydata

The command import ecg should now work for any Python scripts or Jupyter Notebooks created and stored in the top-level ecg directory (i.e. ecgHub/ecg). Files not used by ecg or generated with ecg, but which are relevant or occasionally needed in scripts which import ecg, can then be stored in the ecgHub folder. (manuscripts, notes, templates, auxiliary csvs, etc.)

Local Installation

To install locally, if you haven't navigated to the package directory:

pip install -e /path/to/package

To install locally, if you're in the package directory:

pip install -e .

The -e flag indicates a symlink, and forces the package to upgrade whenever the source directory changes (e.g. if you pull from github)

Downloading KEGG data (`kegg.py`)

Downloading and formatting KEGG data can be done through by importing the ecg package in a script, or through a command line interface (CLI).

Using import

Downloading and running pipeline

import ecg

path = "mydata/kegg"
kegg = ecg.Kegg(path)
kegg.download(run_pipeline=True,dbs=["pathway","enzyme","reaction","compound"]) ## These are the default arguments

Updating KEGG directory with latest entries

Note: Updating will NOT reflect changes made to invdividual entries' fields, and it will NOT remove entries which have been removed from KEGG. It will only add entries which have been added. To guarantee the most up-to-date KEGG database, a full re-download is necessary.

import ecg

path = "mydata/kegg"
K = ecg.Kegg(path)
K.update(metadata=True) ## These are the default arguments

## Built-in public methods
K.download;
K.update;
K.path;
K.version;
K.lists;

Using CLI

Note: If more than one user supplied database is provided it must be done so with its own flag. Each flag can only accept one argument.

Example: python kegg.py --path mydir --db compound reaction

"""
usage: kegg.py [-h] [--rp RP]
               [--db {pathway,brite,module,ko,genome,<org>,vg,ag,compound,glycan,reaction,rclass,enzyme,network,hsa_var,disease,drug,dgroup,environ} [{pathway,brite,module,ko,genome,<org>,vg,ag,compound,glycan,reaction,rclass,enzyme,network,hsa_var,disease,drug,dgroup,environ} ...]]
               [--md MD] --path PATH [--download DOWNLOAD] [--update UPDATE]

Retrieve KEGG databases and format them for use in network expansions.

optional arguments:
  -h, --help            show this help message and exit
  --rp RP               Whether or not to run the full pipline. (Default =
                        True)
  --db {pathway,brite,module,ko,genome,<org>,vg,ag,compound,glycan,reaction,rclass,enzyme,network,hsa_var,disease,drug,dgroup,environ} [{pathway,brite,module,ko,genome,<org>,vg,ag,compound,glycan,reaction,rclass,enzyme,network,hsa_var,disease,drug,dgroup,environ} ...]
                        Databases to download. For more information on dbs see
                        KEGG DB Links. (Default = ["pathway", "enzyme",
                        "reaction", "compound"])
  --md MD               Whether to add metadata fields from "RXXXXX.json" into
                        master.json. (Default = True)
  --path PATH           Directory where KEGG will be downloaded to or updated.
                        (Required)
  --download DOWNLOAD   Whether to download KEGG and run pipeline to format
                        data. (Default = True)
  --update UPDATE       Whether or not to update existing KEGG directory.
                        Note: Updating will NOT reflect changes made to
                        invdividual entry fields, and it will NOT remove
                        entries which have been removed from KEGG. It will
                        only add entries which have been added. To guarantee
                        the most up-to-date KEGG database, a full re-download
                        is necessary. (Default = False)
"""

Output format

The default file structure output from Kegg.download() looks like:

mydata/kegg
|-entries
|  |-compounds 
|  |-reactions 
|  |-enzymes 
|  |-pathways 
|
|-lists
|  |-compounds 
|  |-reactions 
|  |-enzymes 
|  |-pathways 
|
|-links
|  |-compound_enzyme
|  |-reaction_compound
|  |-enzyme_compound
|  |-pathway_reaction
|  ...
|
version.json # file with all the info that is in the version entry of the master, created upon download and updated upon update
|
master.json

The defualt structure of the version.json file looks approximately like:

Kegg.version  #returns info from http://rest.kegg.jp/info/kegg
|- Kegg.version["original"] = dict() # KEGG stats at time of initial retrieval
|    |-...["release_short"] = float # returns release of database eg. 90.0
|    |-...["release_long"] = str # returns 90.0+/06-06, Jun 19
|    |-...["count_info"] = dict() # returns number of entries in each of the subcategories as found on KEGG's info page at time of retrieval
|        |-...["pathway"] = int
|        |-...["compound"] = int
|        |-...["reaction"] = int
|        |-...["enzyme"] = int
|    |-...["count_lists"] = dict() # returns number of entries in each of the subcategories as found by counting the items in Kegg.version["original"]["lists"]
|        |-...["pathway"] = int
|        |-...["compound"] = int
|        |-...["reaction"] = int
|        |-...["enzyme"] = int
|    |-...["lists"] = dict() # returns name of each entry in each of the subcategories
|        |-...["pathway"] = list()
|        |-...["compound"] = list()
|        |-...["reaction"] = list()
|        |-...["enzyme"] = list()
|- Kegg.version["updates"] = list(dict(),dict(),...) # KEGG stats after each update
|- Kegg.version["current"] = dict() # KEGG stats of latest update

Downloading JGI data (`jgi.py`)

Downloading JGI data can be done through by importing the ecg package in a script, or through a command line interface (CLI).

Using import

Downloading and running pipeline

from ecg import jgi
import os

chromedriver_path = os.path.expanduser("~")+"/chromedriver" # "~/chromedriver" should also work
path = "mydata/jgi"
domain = "Eukaryota"

J = jgi.Jgi()
J.scrape_domain(path, 
                domain,
                database='all', 
                assembly_types = ['assembled','unassembled','both'])

## Built-in public methods
J.scrape_domain;
J.scrape_urls; # argument organism_urls should be a list of full urls
J.homepage_url;
J.driver;

Using CLI

Example: python jgi.py --cd_path=/usr/bin/chromedriver --path myjgidir --domain Bacteria --db=jgi

"""
usage: jgi.py [-h] --path PATH
              [--scrape_domain SCRAPE_DOMAIN | --scrape_urls SCRAPE_URLS]
              [--organism_urls ORGANISM_URLS [ORGANISM_URLS ...]] --domain
              {Eukaryota,Bacteria,Archaea,*Microbiome,Plasmids,Viruses,GFragment,cell,sps,Metatranscriptome}
              [--cd_path CD_PATH] [--hp_url HP_URL] [--db {jgi,all}]
              [--at {assembled,unassembled,both} [{assembled,unassembled,both} ...]]

Retrieve enzyme data from JGI genomes and metagenomes.

optional arguments:
  -h, --help            show this help message and exit
  --path PATH           Directory where JGI data will be downloaded to.
                        (Required)
  --scrape_domain SCRAPE_DOMAIN
                        Download an entire JGI domain and run pipeline to
                        format data (Default = True).
  --scrape_urls SCRAPE_URLS
                        Download data from one or more (meta)genomes by URL.
                        (Default = False).
  --organism_urls ORGANISM_URLS [ORGANISM_URLS ...]
                        List of (meta)genomes by URL for scrape_urls.
                        (Required only if scrape_urls == True)
  --domain {Eukaryota,Bacteria,Archaea,*Microbiome,Plasmids,Viruses,GFragment,cell,sps,Metatranscriptome}
                        JGI valid domain to scrape data from. (Required)
  --cd_path CD_PATH     Path pointing to chromedriver executable. (Required)
  --hp_url HP_URL       URL of JGI's homepage. (Optional. Default =
                        https://img.jgi.doe.gov/cgi-bin/m/main.cgi)
  --db {jgi,all}        To use only JGI annotated organisms or all organisms.
                        (Optional. Default = all)
  --at {assembled,unassembled,both} [{assembled,unassembled,both} ...]
                        Assembly types. Only used for metagenomic domains.
                        Ignored for others. (Optional. Default = [assembled,
                        unassembled, both])
"""

Output format

The default file structure output from jgi.Jgi().scrape_domain("myjgidir","Eukarayota") looks like:

mydata/jgi
|-Eukarayota
|  |-combined_taxon_ids 
|  |-missing_enzymes.json 
|  |-taxon_ids
|    |-2789789765.json
|    |-2789789766.json
|    ...
|-Bacteria
|  ...

Filtering JGI data

We apply the following criteria for filtering EC data from JGI:

Remove enzymes with partial names (example: 1.1.1.-)
Remove enzymes whose 1st digit is '7'
Remove enzymes not found in KEGG (these are apparently associated with glycans)
Remove fungal annotated entries in eukaryotes as they are suspect
Remove entires that are at extremes of percent functional coding genes (<10% or >90%)
Remove entries that are below the minimal gene size: 1364 for prokaryotes, 4718 for eukaryotes, and 27280 for metagenomes

See Gagler et al. 2021 for more details.

Use jgi_pipeline_create_csvs.py followed by filter_raw.py to get clean data.

These scripts were previously in Bradley Karas' repository. They were titled jgi_pipeline.py and filter_raw.py, respectively. The only changes that have been made between these sets of scripts are the paths, and changing one of the keys from the metadata from "Lineage" to "NCBI Taxonomy Lineage" ("Lineage" as a key no longer existed in the JGI metadata, hence the update). We made this change in the jgi_pipeline... script.

Getting biosystem reaction lists and network graphs using KEGG and JGI (`ecg.py`)

Using import

Writing reaction jsons

In preparation for doing network expansions using BioXP

from ecg import ecg

E = ecg.Ecg()
ec_rxn_link_json = "mydata/kegg/links/enzyme_reaction.json"

## Write reactions from one jgi biosystem file
biosystem_json = "mydata/jgi/Eukaryota/taxon_ids/2789789765.json"
E.write_biosystem_rxns(biosystem_json,
                       ec_rxn_link_json,
                       outdir="mydata/jgi/Eukaryota/taxon_reactions")

## Write reactions from all jgi biosystem files in the directory
biosystem_json_dir = "mydata/jgi/Eukaryota/taxon_ids"
E.write_biosystem_rxns(biosystem_json_dir,
                       ec_rxn_link_json,
                       outdir="mydata/jgi/Eukaryota/taxon_reactions")

Writing graphs

Note: You must generate biosystem reaction jsons first

Not required for network expansions using BioXP

from ecg import ecg

E = ecg.Ecg()
master_json = "mydata/kegg/master.json"

## Write graphs from one biosystem rxn file
biosystem_rxn_json = "mydata/jgi/Eukaryota/taxon_reactions/2789789765.json"
E.write_biosystem_graphs(biosys_rxn_json,
                        master_json,
                        graphtypes=["unipartite-undirected-subfromdirected"],
                        outdir="mydata/jgi/Eukaryota/graphs",
                        missingdir="mydata/jgi/Eukaryota/taxon_with_rxns_missing_from_kegg",
                        verbose=True)

## Write graphs from all biosystem rxn files in the directory
biosystem_rxn_json_dir = "mydata/jgi/Eukaryota/taxon_reactions"
E.write_biosystem_graphs(biosys_rxn_json,
                        master_json,
                        graphtypes=["unipartite-undirected-subfromdirected"],
                        outdir="mydata/jgi/Eukaryota/graphs",
                        missingdir="mydata/jgi/Eukaryota/taxon_with_rxns_missing_from_kegg",
                        verbose=True)

## Built-in public methods
E.write_biosystem_rxns;
E.write_biosystem_graphs

Note: Implemented graphtypes include:

bipartite-directed-rxnsub
bipartite-undirected-rxnsub
unipartite-undirected-rxn
unipartite-directed-sub
unipartite-undirected-sub
unipartite-undirected-subfromdirected #same connection rules used in "Universal Scaling" paper

Using CLI

Example: python ecg.py --jgi=mydata/jgi/Eukaryota/taxon_reactions/2789789765.json --master= mydata/kegg/master.json --graphtypes unipartite-undirected-subfromdirected --outdir mydata/jgi/Eukaryota/taxon_reactions --graphoutdir mydata/jgi/Eukaryota/graphs --missingdir mydata/jgi/Eukaryota/taxon_with_rxns_missing_from_kegg

"""
usage: ecg.py [-h] --jgi JGI --ecrxn ECRXN --biorxn BIORXN --master MASTER
              [--write_rxns WRITE_RXNS] [--write_graphs WRITE_GRAPHS]
              [--outdir OUTDIR]
              [--graphtypes {bipartite-directed-rxnsub,bipartite-undirected-rxnsub,unipartite-undirected-rxn,unipartite-directed-sub,unipartite-undirected-sub,unipartite-undirected-subfromdirected} [{bipartite-directed-rxnsub,bipartite-undirected-rxnsub,unipartite-undirected-rxn,unipartite-directed-sub,unipartite-undirected-sub,unipartite-undirected-subfromdirected} ...]]
              [--graphoutdir GRAPHOUTDIR] [--missingdir MISSINGDIR]
              [--verbose VERBOSE]

WARNING. CLI HAS NOT BEEN IMPLEMENTED OR TESTED YET. Combine KEGG derived
reaction data with JGI derived enzyme data to generate reaction lists
(meta)genomes.

optional arguments:
  -h, --help            show this help message and exit
  --jgi JGI             Filepath to the directory or file where JGI data is
                        located. (Required)
  --ecrxn ECRXN         Filepath to "enzyme_reaction.json"; the json
                        containing relationship between ec numbers and
                        reactions. (Required)
  --biorxn BIORXN       Filepath to the biosystem reaction json file.
                        (Required)
  --master MASTER       Filepath to "master.json"; json with information
                        details about all KEGG reactions. (Required)
  --write_rxns WRITE_RXNS
                        Write reaction lists from either a single biosystem
                        file or biosystem directory; all JGI jsons. (Default =
                        True)
  --write_graphs WRITE_GRAPHS
                        Write gmls from either a single biosystem reactionfile
                        or biosystem reaction directory; all JGI reaction
                        jsons. (Default = True)
  --outdir OUTDIR       Path where biosystem reactions will be saved. (Default
                        = "taxon_reactions")
  --graphtypes {bipartite-directed-rxnsub,bipartite-undirected-rxnsub,unipartite-undirected-rxn,unipartite-directed-sub,unipartite-undirected-sub,unipartite-undirected-subfromdirected} [{bipartite-directed-rxnsub,bipartite-undirected-rxnsub,unipartite-undirected-rxn,unipartite-directed-sub,unipartite-undirected-sub,unipartite-undirected-subfromdirected} ...]
                        Which types of graphs to write to gml files. (Default
                        = "unipartite-undirected-subfromdirected")
  --graphoutdir GRAPHOUTDIR
                        The directory to store subdirs for each graph type,
                        and subsequent gml files. (Default = "graphs")
  --missingdir MISSINGDIR
                        The directory to store reactions which are missing
                        from biosystems as jsons. (Default =
                        "taxon_with_rxns_missing_from_kegg")
  --verbose VERBOSE     If True, prints the graph types as they are created.
                        (Default = True)
"""

Output format

Running ecg.Ecg() as intended (and with default arguments) will generate the following new folders:

graphs
taxon_reactions
taxon_with_rxns_missing_from_kegg (directory is only created if your taxon_ids contain enzymes with reactions which are missing from the KEGG database)

So your file structure will now look like this:

myjgidir
|-Eukarayota
|  |-combined_taxon_ids 
|  |-graphs
|     |-bipartite-directed-rxnsub
|        |-2789789765.gml
|        |-2789789766.gml
|        ...
|     |-bipartite-directed-rxnsub
|        |-2789789765.gml
|        |-2789789766.gml
|        ...
|     ...
|  |-missing_enzymes.json 
|  |-taxon_ids
|     |-2789789765.json
|     |-2789789766.json
|     ...
|  |-taxon_reactions
|     |-2789789765.json
|     |-2789789766.json
|     ...
|  |-taxon_with_rxns_missing_from_kegg
|     |-9999999999.json
|     ...
|-Bacteria
|  ...

Name		Name	Last commit message	Last commit date
Latest commit History 179 Commits
.images		.images
ecg.egg-info		ecg.egg-info
ecg		ecg
script		script
test		test
.gitignore		.gitignore
LICENSE		LICENSE
README.md		README.md
Untitled.ipynb		Untitled.ipynb
setup.py		setup.py

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Repository files navigation

ecg

Dependencies

Quickstart

Local Installation

Downloading KEGG data (`kegg.py`)

Using import

Downloading and running pipeline

Updating KEGG directory with latest entries

Using CLI

Output format

Downloading JGI data (`jgi.py`)

Using import

Downloading and running pipeline

Using CLI

Output format

Filtering JGI data

Getting biosystem reaction lists and network graphs using KEGG and JGI (`ecg.py`)

Using import

Writing reaction jsons

Writing graphs

Using CLI

Output format

About

Releases

Packages

Contributors 6

Languages

License

ELIFE-ASU/ecg

Folders and files

Latest commit

History

Repository files navigation

ecg

Dependencies

Quickstart

Local Installation

Downloading KEGG data (kegg.py)

Using import

Downloading and running pipeline

Updating KEGG directory with latest entries

Using CLI

Output format

Downloading JGI data (jgi.py)

Using import

Downloading and running pipeline

Using CLI

Output format

Filtering JGI data

Getting biosystem reaction lists and network graphs using KEGG and JGI (ecg.py)

Using import

Writing reaction jsons

Writing graphs

Using CLI

Output format

About

Resources

License

Stars

Watchers

Forks

Releases

Packages 0

Contributors 6

Languages

Downloading KEGG data (`kegg.py`)

Downloading JGI data (`jgi.py`)

Getting biosystem reaction lists and network graphs using KEGG and JGI (`ecg.py`)

Packages