1. Data cleaning with the Kurator toolkit
Bridging the gap between conventional scripting and
high-performance workflow automation
Timothy McPhillips, David Lowery, James Hanken,
Bertram Ludäscher, James A. Macklin, Paul J. Morris,
Robert A. Morris, Tianhong Song, and John Wieczorek
TDWG 2015 - Biodiversity Informatics Services and Workflows Symposium
Nairobi, Kenya - September 30, 2015

2. Kurator: workflow automation for
cleaning biodiversity data
Project aims
§ Facilitate cleaning of biodiversity data.
§ Support both traditional scripting and
high-­performance scientific workflows.
§ Deliver much more than a fixed set of
configurable workflows.
Technical strategy
§ Wrap Akka actor toolkit in a curation-­oriented
scientific workflow language and runtime.
§ Enable scientists who write scripts to add their
own new data validation and cleaning actors.
§ Bring to scripts major advantages of workflow
automation: prospective and retrospective
provenance.
§ Bridge gaps between data validation services,
data cleaning scripts, and pipelined data curation
workflows.
Empower users and developers of scripts, actors, and workflows.

3. What some of us think of when we hear
the term ‘scientific workflows’
Phylogenetics workflow in Kepler (2005)
Graphical interface
§ Canvas for assembling and
displaying the workflow.
§ Library of workflow blocks
(‘actors’) that can be
dragged onto the canvas
and connected.
§ Arrows that represent
control dependencies or
paths of data flow.
§ A run button.
These features are not
essential to managing
actual scientific
workflows.

4. 10 essential functions
of a scientific workflow system
1. Automate programs and services scientists already use.
2. Schedule invocations of programs and services correctly and efficiently
– in parallel where possible.
3. Manage data flow to, from, and between programs and services.
4. Enable scientists (not just developers) to author or modify workflows easily.
5. Predict what a workflow will do when executed: prospective provenance.
6. Record what actually happens during workflow execution.
7. Reveal retrospective provenance – how workflow products were derived
from inputs via programs and services.
8. Organize intermediate and final data products as desired by users.
9. Enable scientists to version, share and publish their workflows.
10. Empower scientists who wish to automate additional programs and
services themselves.
These functions–not actors—distinguish scientific workflow
automation from general scientific software development.

5. Why build yet another system?
Available systems
§ Kepler (Ptolemy II), Taverna, VisTrails…
§ Familiar graphical programming
environments.
Limitations
§ Often little support for organizing
intermediate and final data products in
ways familiar to scientists.
§ Professional software developers
frequently are needed to develop new
components or workflows.
Huge gap between how these systems
are used and how scientists already
automate their analyses—via scripting.
Part of a Kepler workflow for
inferring phylogenetic trees from
protein sequences.

6. Avoiding actor ‘overuse injuries’
Overuse the actor paradigm…
§ In many systems workflows can be reused as
actors, or ‘subworkflows’ in other workflows.
§ This is a necessary abstraction when
workflow systems are not well-­integrated
with scripting languages.
§ Each actor at left is a page of Java code. But
this whole ‘subworkflow’ could be written in a
half a page of Python!
…or use the right tools for the right job!
§ In Kurator we want to enable the actor
abstraction where it pays off the most—as
the unit of parallelism.
§ For specifying behavior inside of actors why
not use easy to understand scripts?
New actors and workflows must be
easy and fast to develop.
Part of a Kepler workflow for
inferring phylogenetic trees from
protein sequences.

7. Data curation workflow using Kepler
FilteredPush explored using workflows for data cleaning
§ First used COMAD workflow model supported by Kepler.
§ Enabled graphical assembly and configuration of custom workflows
from library of actors.
Highlighted potential performance limitations of workflow engines.

8. FP-­Akka workflows
load data
check scientific name
check basis of record
check date collected
check lat/long
write out results
§ FilteredPush next investigated use of
the Akka actor toolkit and platform.
§ Widely used in industry, well
supported, and rapidly advancing.
§ Efficient parallel execution across
multiple processors and compute nodes.
§ Improved performance and scalability
compared to Kepler.
Limitations of directly using Akka
§ Writing new Akka actors and programs
requires Java (or Scala) experience.
§ Must address many parallel
programming challenges from scratch.
Advanced programming skills
required to write Akka programs
(‘workflows’) that run correctly.

9. Akka partly supports two essential
workflow platform requirements
The Kurator toolkit will satisfy the rest as needed.
1. Automate programs and services scientists already use.
2. Schedule invocations of programs and services correctly and
efficiently–in parallel where possible.
3. Manage data flow to, from, and between programs and services.
4. Enable scientists (not just developers) to author or modify workflows easily.
5. Predict what a workflow will do when executed: prospective provenance.
6. Record what actually happens during workflow execution.
7. Reveal retrospective provenance – how workflow products were derived from
inputs via programs and services.
8. Organize intermediate and final data products as desired by users.
9. Enable scientists to version, share and publish their workflows.
10. Empower scientists who wish to automate additional programs and services
themselves.

10. The Kurator Toolkit
YesWorkflow (YW)
§ Add YW annotations to any script or program
that supports text comments. Highlight the
workflow structure in the script.
§ Visualize or query prospective provenance
before running a script.
§ Reconstruct, visualize, and query retrospective
provenance after running the script.
§ Integrate provenance gathered from file names
and paths, log files, data file headers, run
metadata, and records of run-­time events.
Kurator-­Akka
§ Write functions or classes in Python or Java. Mark up with YesWorkflow.
§ Declare how scripts or Java code can be used as actors. Short YAML blocks.
§ Declare workflows. List actors and specify their connections. More YAML.
§ Run workflow. Use Akka for parallelization transparently—and correctly.
§ Reconstruct retrospective provenance of workflow products.

11. Example: data validation and
cleaning using WoRMS web services
1) Write simple Python functions (or a class) that wrap the web services provided
by the World Register of Marine Species (WoRMS).
2) Develop a Python script that uses the service wrapper functions from (1) to
clean a set of records provided in CSV format.
3) Mark up the script written in (2) with YesWorkflow annotations and
graphically display the script as a workflow.
4) Factor out of (2) a Python function that can clean a single record using the
WoRMS wrapper functions in (1).
5) Write a block of YAML that declares how the record-­cleaning function in (4) can
be used as an actor in a Kurator-­Akka workflow.
6) Declare using YAML a workflow that uses the actor declared in (5) along with
actors for reading and writing CSV files.
7) Run the workflow (6) on a sample data set with the CSV reader, CSV writer
and WoRMS validation actors all running in parallel.
README at https://github.com/kurator-­org/kurator-­validation shows how to:
Illustrates how Kurator aims to facilitate composition of actors and
high-­performance workflows from simple functions and scripts.

12. class WoRMSService(object):
"""
Class for accessing the WoRMS taxonomic name database via the AphiaNameService.
The Aphia names services are described at http://marinespecies.org/aphia.php?p=soap.
"""
WORMS_APHIA_NAME_SERVICE_URL = 'http://marinespecies.org/aphia.php?p=soap&wsdl=1’
def __init__(self, marine_only=False):
""" Initialize a SOAP client using the WSDL for the WoRMS Aphia names service"""
self._client = Client(self.WORMS_APHIA_NAME_SERVICE_URL)
self._marine_only = marine_only
def aphia_record_by_exact_taxon_name(self, name):
"""
Perform an exact match on the input name against the taxon names in WoRMS.
This function first invokes an Aphia names service to lookup the Aphia ID for
the taxon name. If exactly one match is returned, this function retrieves the
Aphia record for that ID and returns it.
"""
aphia_id = self._client.service.getAphiaID(name, self._marine_only);
if aphia_id is None or aphia_id == -­‐999: # -­‐999 indicates multiple matches
return None
else:
return self._client.service.getAphiaRecordByID(aphia_id)
def aphia_record_by_fuzzy_taxon_name(self, name):
:
:
Python class wrapping WoRMS services
WoRMS services

13. # @BEGIN find_matching_worms_record
# @IN original_scientific_name
# @OUT matching_worms_record
# @OUT worms_lsid
worms_match_result = None
worms_lsid = None
# first try exact match of the scientific name against WoRMS
matching_worms_record = worms.aphia_record_by_exact_taxon_name(original_scientific_name)
if matching_worms_record is not None:
worms_match_result = 'exact'
# otherwise try a fuzzy match
else:
matching_worms_record = worms.aphia_record_by_fuzzy_taxon_name(original_scientific_name)
if matching_worms_record is not None:
worms_match_result = 'fuzzy’
# if either match succeeds extract the LSID for the taxon
if matching_worms_record is not None:
worms_lsid = matching_worms_record['lsid']
# @END find_matching_worms_record
Finding matching WoRMS records
in the data cleaning script
YesWorkflow annotation
marking start of workflow step.
End of workflow step.
Variables serving as inputs and
outputs to workflow step.
Calls to WoRMS service
wrapper functions.

14. YW rendering of script
Workflow steps each
delimited by @BEGIN
and @END annotations
in script.
Data flowing into and out of
find_matching_worms_record
workflow step.
YesWorkflow infers connections
between workflow steps and what
data flows through them by matching
@IN and @OUT annotations.

15. # @BEGIN compose_cleaned_record
# @IN original_record
# @IN worms_lsid
# @IN updated_scientific_name
# @IN original_scientific_name
# @IN updated_authorship
# @IN original_authorship
# @OUT cleaned_record
cleaned_record = original_record
cleaned_record['LSID'] = worms_lsid
cleaned_record['WoRMsMatchResult'] = worms_match_result
if updated_scientific_name is not None:
cleaned_record['scientificName'] = updated_scientific_name
cleaned_record['originalScientificName'] = original_scientific_name
if updated_authorship is not None:
cleaned_record['scientificNameAuthorship'] = updated_authorship
cleaned_record['originalAuthor'] = original_authorship
# @END compose_cleaned_record
The compose_cleaned_record step in
the data cleaning script

16. A function for cleaning one record
def curate_taxon_name_and_author(self, input_record):
# look up record for input taxon name in WoRMS taxonomic database
is_exact_match, aphia_record = (
self._worms.aphia_record_by_taxon_name(input_record['TaxonName']))
if aphia_record is not None:
# save taxon name and author values from input record in new fields
input_record['OriginalName'] = input_record['TaxonName']
input_record['OriginalAuthor'] = input_record['Author']
# replace taxon name and author fields in input record with values in aphia record
input_record['TaxonName'] = aphia_record['scientificname']
input_record['Author'] = aphia_record['authority']
# add new fields
input_record['WoRMsExactMatch'] = is_exact_match
input_record['lsid'] = aphia_record['lsid’]
else:
input_record['OriginalName'] = None
input_record['OriginalAuthor'] = None
input_record['WoRMsExactMatch'] = None
input_record['lsid'] = None
return input_record
Factoring out the core functionality
of a script into a reusable function
is a natural step in script evolution.

17. Declaring the function as an actor
-­‐ id: WoRMSNameCurator
type: PythonClassActor
properties:
pythonClass: org.kurator.validation.actors.WoRMSCurator.WoRMSCurator
onData: curate_taxon_name_and_author
Actor type identifier
referenced when composing a
workflow that uses the actor.
Python class declaring
the function as a
method (optional).
Name of Python function called
for each data item received by
the actor at run time.
Besides this block of YAML, no additional code needs
to be written to convert the function into an actor.

18. Simple workflow using the new actor
components:
-­‐ id: ReadInput
type: CsvFileReader
-­‐ id: CurateRecords
type: WoRMSNameCurator
properties:
listensTo:
-­‐ !ref ReadInput
-­‐ id: WriteOutput
type: CsvFileWriter
properties:
listensTo:
-­‐ !ref CurateRecords
-­‐ id: WoRMSNameValidationWorkflow
type: Workflow
properties:
actors:
-­‐ !ref ReadInput
-­‐ !ref CurateRecords
-­‐ !ref WriteOutput
Actor that reads input from a CSV
file. Emits records one at a time.
The WoRMSNameCurator actor
declared on the previous slide.
Processes each received record in turn.
Actor that writes received records
to an output CSV file one by one.
The listensTo property is used to
declare how actors are connected.
Declaration of workflow as a
composition of the three actors.
Actors run concurrently, each working on
different records at the same time, when the
workflow is executed by Kurator-­Akka.

19. CsvFileReader
WoRMSNameCurator
CsvFileWriter
Correspondence of
steps in script to
actors in workflow

20. Running the workflow
$ ka -­‐f WoRMS_name_validation.yaml < WoRMS_name_validation_input.csv
ID,TaxonName,Author,OriginalName,OriginalAuthor,WoRMsExactMatch,lsid
37929,Architectonica reevei,"(Hanley, 1862)",Architectonica
reevi,,false,urn:lsid:marinespecies.org:taxname:588206
37932,Rapana rapiformis,"(Born, 1778)",Rapana rapiformis,"(Von Born,
1778)",true,urn:lsid:marinespecies.org:taxname:140415
180593,Buccinum donomani,"(Linnaeus, 1758)",,,,
179963,Codakia paytenorum,"(Iredale, 1937)",Codakia paytenorum,"Iredale,
1937",true,urn:lsid:marinespecies.org:taxname:215841
0,Rissoa venusta,"Garrett, 1873",Rissoa
venusta,,true,urn:lsid:marinespecies.org:taxname:607233
62156,Rissoa venusta,"Garrett, 1873",Rissoa
venusta,Phil.,true,urn:lsid:marinespecies.org:taxname:607233
$
Workflow can be run
at the command line.
Actors can be created from simple scripts, and workflows can be
run like scripts. Simple YAML files wire everything together.
Actors can read and write standard
input and output like any script.

21. The road ahead
The immediate future
§ YesWorkflow support for graphically rendering Kurator-­Akka workflows.
§ Combining prospective and retrospective provenance from workflow
declarations and YesWorkflow-­annotated scripts used as actors.
§ Enhancements to YAML workflow declarations enabling Akka support for
creating and managing multiple instances of each actor for higher
throughput.
§ Exploration of advanced Akka features in FP-­Akka framework, followed
by generalization of these features in Kurator-­Akka so that all users can
benefit.
Future possibilities
§ Support for additional actor scripting languages (e.g., R).
§ Actor function wrappers for other workflow and dataflow systems.
§ Graphical user interface for composing and running workflows.
§ Your suggestions?