diff options
| -rw-r--r-- | README.md | 21 | ||||
| -rw-r--r-- | bh20seqanalyzer/main.py | 148 | ||||
| -rw-r--r-- | bh20sequploader/bh20seq-schema.yml | 89 | ||||
| -rw-r--r-- | bh20sequploader/main.py | 16 | ||||
| -rw-r--r-- | bh20sequploader/qc_metadata.py | 6 | ||||
| -rw-r--r-- | doc/DEVELOPMENT.md | 7 | ||||
| -rw-r--r-- | doc/INSTALL.md | 31 | ||||
| -rw-r--r-- | example/metadata.yaml | 49 | ||||
| -rw-r--r-- | example/minimal_example.yaml | 14 | ||||
| -rw-r--r-- | paper/paper.bib | 0 | ||||
| -rw-r--r-- | paper/paper.md | 110 |
11 files changed, 401 insertions, 90 deletions
diff --git a/README.md b/README.md index 1448f4c..1c6f239 100644 --- a/README.md +++ b/README.md @@ -122,19 +122,7 @@ It should print some instructions about how to use the uploader. ## Installation with GNU Guix -Another way to install this tool is inside a [GNU Guix Environment](https://guix.gnu.org/manual/en/html_node/Invoking-guix-environment.html), which can handle installing dependencies for you even when you don't have root access on an Ubuntu system. - -1. **Set up and enter a container with the necessary dependencies.** After installing Guix as `~/opt/guix/bin/guix`, run: - -```sh -~/opt/guix/bin/guix environment -C guix --ad-hoc git python openssl python-pycurl nss-certs -``` - -2. **Install the tool.** From there you can follow the [user installation instructions](#installation-with-pip3---user). In brief: - -```sh -pip3 install --user git+https://github.com/arvados/bh20-seq-resource.git@master -``` +For running/developing the uploader with GNU Guix see [INSTALL.md](./doc/INSTALL.md) # Usage @@ -148,7 +136,7 @@ bh20-seq-uploader example/sequence.fasta example/metadata.json All these uploaded sequences are being fed into a workflow to generate a [pangenome](https://academic.oup.com/bib/article/19/1/118/2566735) for the virus. You can replicate this workflow yourself. -Get your SARS-CoV-2 sequences from GenBank in `seqs.fa`, and then run: +An example is to get your SARS-CoV-2 sequences from GenBank in `seqs.fa`, and then run a series of commands ```sh minimap2 -cx asm20 -X seqs.fa seqs.fa >seqs.paf @@ -157,6 +145,7 @@ odgi build -g seqs.gfa -s -o seqs.odgi odgi viz -i seqs.odgi -o seqs.png -x 4000 -y 500 -R -P 5 ``` -For more information on building pangenome models, [see this wiki page](https://github.com/virtual-biohackathons/covid-19-bh20/wiki/Pangenome#pangenome-model-from-available-genomes). - +Here we convert such a pipeline into the Common Workflow Language (CWL) and +sources can be found [here](https://github.com/hpobio-lab/viral-analysis/tree/master/cwl/pangenome-generate). +For more information on building pangenome models, [see this wiki page](https://github.com/virtual-biohackathons/covid-19-bh20/wiki/Pangenome#pangenome-model-from-available-genomes). diff --git a/bh20seqanalyzer/main.py b/bh20seqanalyzer/main.py index 78e32c9..2030c1e 100644 --- a/bh20seqanalyzer/main.py +++ b/bh20seqanalyzer/main.py @@ -13,21 +13,30 @@ logging.basicConfig(format="[%(asctime)s] %(levelname)s %(message)s", datefmt="% level=logging.INFO) logging.getLogger("googleapiclient.discovery").setLevel(logging.WARN) -def validate_upload(api, collection, validated_project): +def validate_upload(api, collection, validated_project, + fastq_project, fastq_workflow_uuid): col = arvados.collection.Collection(collection["uuid"]) # validate the collection here. Check metadata, etc. valid = True - if "sequence.fasta" not in col: - valid = False - logging.warn("Upload '%s' missing sequence.fasta", collection["name"]) if "metadata.yaml" not in col: logging.warn("Upload '%s' missing metadata.yaml", collection["name"]) valid = False else: metadata_content = ruamel.yaml.round_trip_load(col.open("metadata.yaml")) - valid = qc_metadata(metadata_content) and valid + #valid = qc_metadata(metadata_content) and valid + if not valid: + logging.warn("Failed metadata qc") + + if valid: + if "sequence.fasta" not in col: + if "reads.fastq" in col: + start_fastq_to_fasta(api, collection, fastq_project, fastq_workflow_uuid) + return False + else: + valid = False + logging.warn("Upload '%s' missing sequence.fasta", collection["name"]) dup = api.collections().list(filters=[["owner_uuid", "=", validated_project], ["portable_data_hash", "=", col.portable_data_hash()]]).execute() @@ -39,7 +48,9 @@ def validate_upload(api, collection, validated_project): if valid: logging.info("Added '%s' to validated sequences" % collection["name"]) # Move it to the "validated" project to be included in the next analysis - api.collections().update(uuid=collection["uuid"], body={"owner_uuid": validated_project}).execute() + api.collections().update(uuid=collection["uuid"], body={ + "owner_uuid": validated_project, + "name": "%s (%s)" % (collection["name"], time.asctime(time.gmtime()))}).execute() else: # It is invalid, delete it. logging.warn("Deleting '%s'" % collection["name"]) @@ -47,28 +58,15 @@ def validate_upload(api, collection, validated_project): return valid -def start_analysis(api, - analysis_project, - workflow_uuid, - validated_project): +def run_workflow(api, parent_project, workflow_uuid, name, inputobj): project = api.groups().create(body={ "group_class": "project", - "name": "Pangenome analysis", - "owner_uuid": analysis_project, + "name": name, + "owner_uuid": parent_project, }, ensure_unique_name=True).execute() - validated = arvados.util.list_all(api.collections().list, filters=[["owner_uuid", "=", validated_project]]) - with tempfile.NamedTemporaryFile() as tmp: - inputobj = { - "inputReads": [] - } - for v in validated: - inputobj["inputReads"].append({ - "class": "File", - "location": "keep:%s/sequence.fasta" % v["portable_data_hash"] - }) tmp.write(json.dumps(inputobj, indent=2).encode('utf-8')) tmp.flush() cmd = ["arvados-cwl-runner", @@ -83,32 +81,102 @@ def start_analysis(api, if comp.returncode != 0: logging.error(comp.stderr.decode('utf-8')) + return project + + +def start_fastq_to_fasta(api, collection, + analysis_project, + fastq_workflow_uuid): + newproject = run_workflow(api, analysis_project, fastq_workflow_uuid, "FASTQ to FASTA", { + "fastq_forward": { + "class": "File", + "location": "keep:%s/reads.fastq" % collection["portable_data_hash"] + }, + "metadata": { + "class": "File", + "location": "keep:%s/metadata.yaml" % collection["portable_data_hash"] + }, + "ref_fasta": { + "class": "File", + "location": "keep:ffef6a3b77e5e04f8f62a7b6f67264d1+556/SARS-CoV2-NC_045512.2.fasta" + } + }) + api.collections().update(uuid=collection["uuid"], + body={"owner_uuid": newproject["uuid"]}).execute() + +def start_pangenome_analysis(api, + analysis_project, + pangenome_workflow_uuid, + validated_project): + validated = arvados.util.list_all(api.collections().list, filters=[["owner_uuid", "=", validated_project]]) + inputobj = { + "inputReads": [], + "metadata": [], + "subjects": [] + } + for v in validated: + inputobj["inputReads"].append({ + "class": "File", + "location": "keep:%s/sequence.fasta" % v["portable_data_hash"] + }) + inputobj["metadata"].append({ + "class": "File", + "location": "keep:%s/metadata.yaml" % v["portable_data_hash"] + }) + inputobj["subjects"].append("keep:%s/sequence.fasta" % v["portable_data_hash"]) + run_workflow(api, analysis_project, pangenome_workflow_uuid, "Pangenome analysis", inputobj) + + +def get_workflow_output_from_project(api, uuid): + cr = api.container_requests().list(filters=[['owner_uuid', '=', uuid], + ["requesting_container_uuid", "=", None]]).execute() + if cr["items"] and cr["items"][0]["output_uuid"]: + return cr["items"][0] + else: + return None + def copy_most_recent_result(api, analysis_project, latest_result_uuid): most_recent_analysis = api.groups().list(filters=[['owner_uuid', '=', analysis_project]], order="created_at desc", limit=1).execute() for m in most_recent_analysis["items"]: - cr = api.container_requests().list(filters=[['owner_uuid', '=', m["uuid"]], - ["requesting_container_uuid", "=", None]]).execute() - if cr["items"] and cr["items"][0]["output_uuid"]: - wf = cr["items"][0] + wf = get_workflow_output_from_project(api, m["uuid"]) + if wf: src = api.collections().get(uuid=wf["output_uuid"]).execute() dst = api.collections().get(uuid=latest_result_uuid).execute() if src["portable_data_hash"] != dst["portable_data_hash"]: logging.info("Copying latest result from '%s' to %s", m["name"], latest_result_uuid) api.collections().update(uuid=latest_result_uuid, body={"manifest_text": src["manifest_text"], - "description": "latest result from %s %s" % (m["name"], wf["uuid"])}).execute() + "description": "Result from %s %s" % (m["name"], wf["uuid"])}).execute() break +def move_fastq_to_fasta_results(api, analysis_project, uploader_project): + projects = api.groups().list(filters=[['owner_uuid', '=', analysis_project], + ["properties.moved_output", "!=", True]], + order="created_at desc",).execute() + for p in projects["items"]: + wf = get_workflow_output_from_project(api, p["uuid"]) + if wf: + logging.info("Moving completed fastq2fasta result %s back to uploader project", wf["output_uuid"]) + api.collections().update(uuid=wf["output_uuid"], + body={"owner_uuid": uploader_project}).execute() + p["properties"]["moved_output"] = True + api.groups().update(uuid=p["uuid"], body={"properties": p["properties"]}).execute() + + def main(): parser = argparse.ArgumentParser(description='Analyze collections uploaded to a project') parser.add_argument('--uploader-project', type=str, default='lugli-j7d0g-n5clictpuvwk8aa', help='') - parser.add_argument('--analysis-project', type=str, default='lugli-j7d0g-y4k4uswcqi3ku56', help='') + parser.add_argument('--pangenome-analysis-project', type=str, default='lugli-j7d0g-y4k4uswcqi3ku56', help='') + parser.add_argument('--fastq-project', type=str, default='lugli-j7d0g-xcjxp4oox2u1w8u', help='') parser.add_argument('--validated-project', type=str, default='lugli-j7d0g-5ct8p1i1wrgyjvp', help='') - parser.add_argument('--workflow-uuid', type=str, default='lugli-7fd4e-mqfu9y3ofnpnho1', help='') - parser.add_argument('--latest-result-uuid', type=str, default='lugli-4zz18-z513nlpqm03hpca', help='') + + parser.add_argument('--pangenome-workflow-uuid', type=str, default='lugli-7fd4e-mqfu9y3ofnpnho1', help='') + parser.add_argument('--fastq-workflow-uuid', type=str, default='lugli-7fd4e-2zp9q4jo5xpif9y', help='') + + parser.add_argument('--latest-result-collection', type=str, default='lugli-4zz18-z513nlpqm03hpca', help='') args = parser.parse_args() api = arvados.api() @@ -116,16 +184,24 @@ def main(): logging.info("Starting up, monitoring %s for uploads" % (args.uploader_project)) while True: + move_fastq_to_fasta_results(api, args.fastq_project, args.uploader_project) + new_collections = api.collections().list(filters=[['owner_uuid', '=', args.uploader_project]]).execute() at_least_one_new_valid_seq = False for c in new_collections["items"]: - at_least_one_new_valid_seq = validate_upload(api, c, args.validated_project) or at_least_one_new_valid_seq + at_least_one_new_valid_seq = validate_upload(api, c, + args.validated_project, + args.fastq_project, + args.fastq_workflow_uuid) or at_least_one_new_valid_seq if at_least_one_new_valid_seq: - start_analysis(api, args.analysis_project, - args.workflow_uuid, - args.validated_project) + start_pangenome_analysis(api, + args.pangenome_analysis_project, + args.pangenome_workflow_uuid, + args.validated_project) - copy_most_recent_result(api, args.analysis_project, args.latest_result_uuid) + copy_most_recent_result(api, + args.pangenome_analysis_project, + args.latest_result_collection) - time.sleep(10) + time.sleep(15) diff --git a/bh20sequploader/bh20seq-schema.yml b/bh20sequploader/bh20seq-schema.yml index 6e0973a..5c962d1 100644 --- a/bh20sequploader/bh20seq-schema.yml +++ b/bh20sequploader/bh20seq-schema.yml @@ -1,36 +1,89 @@ +$base: http://biohackathon.org/bh20-seq-schema +$namespaces: + sch: https://schema.org/ + efo: http://www.ebi.ac.uk/efo/ + obo: http://purl.obolibrary.org/obo/ $graph: -- name: sampleInformationSchema +- name: hostSchema type: record fields: - location: string - host: string - sequenceTechnology: string - assemblyMethod: string + host_species: + type: string + jsonldPredicate: + _id: http://www.ebi.ac.uk/efo/EFO_0000532 + host_id: string + host_common_name: string? + host_sex: string? + host_age: int? + host_age_unit: string? + host_health_status: string? + host_treatment: + type: string? + jsonldPredicate: + _id: http://www.ebi.ac.uk/efo/EFO_0000727 + additional_host_information: string? -- name: InstituteInformationSchema +- name: sampleSchema type: record fields: - OriginatingLab: string - SubmittingLab: string + collector_name: string + collecting_institution: string + specimen_source: string? + collection_date: string? + collection_location: + type: string? + jsonldPredicate: + _id: https://schema.org/fromLocation + sample_storage_conditions: string? + additional_collection_information: string? -- name: SubmitterInformationSchema +- name: virusSchema type: record fields: - Submitter: string - submissionDate: string + virus_species: string? + virus_strain: string? -- name: VirusDetailSchema +- name: technologySchema type: record fields: - VirusName: string - AccessionId: string + sample_sequencing_technology: + type: string + jsonldPredicate: + _id: http://www.ebi.ac.uk/efo/EFO_0000532 + sequence_assembly_method: + type: string? + jsonldPredicate: + _id: http://www.ebi.ac.uk/efo/EFO_0002699 + sequencing_coverage: + type: string? + jsonldPredicate: + _id: http://purl.obolibrary.org/obo/FLU_0000848 + +- name: submitterSchema + type: record + fields: + submitter_name: string + submitter_address: string? + originating_lab: string + lab_address: string? + provider_sample_id: string? + submitter_sample_id: string? + authors: string? + submitter_id: string? - name: MainSchema type: record documentRoot: true fields: - sampleInformation: sampleInformationSchema - InstituteInformation: InstituteInformationSchema - SubmitterInformation: SubmitterInformationSchema - VirusDetail: VirusDetailSchema + host: hostSchema + sample: sampleSchema + virus: virusSchema? + technology: technologySchema + submitter: submitterSchema + sequencefile: + doc: The subject (eg the fasta/fastq file) that this metadata describes + type: string? + jsonldPredicate: + _id: "@id" + _type: "@id" diff --git a/bh20sequploader/main.py b/bh20sequploader/main.py index 8b8fefe..bf74ea5 100644 --- a/bh20sequploader/main.py +++ b/bh20sequploader/main.py @@ -6,7 +6,7 @@ import json import urllib.request import socket import getpass -from .qc_metadata import qc_metadata +import qc_metadata ARVADOS_API_HOST='lugli.arvadosapi.com' ARVADOS_API_TOKEN='2fbebpmbo3rw3x05ueu2i6nx70zhrsb1p22ycu3ry34m4x4462' @@ -20,12 +20,18 @@ def main(): api = arvados.api(host=ARVADOS_API_HOST, token=ARVADOS_API_TOKEN, insecure=True) - qc_metadata(args.metadata.name) + if not qc_metadata(args.metadata.name): + print("Failed metadata qc") + exit(1) col = arvados.collection.Collection(api_client=api) - print("Reading FASTA") - with col.open("sequence.fasta", "w") as f: + if args.sequence.name.endswith("fasta") or args.sequence.name.endswith("fa"): + target = "sequence.fasta" + elif args.sequence.name.endswith("fastq") or args.sequence.name.endswith("fq"): + target = "reads.fastq" + + with col.open(target, "w") as f: r = args.sequence.read(65536) print(r[0:20]) while r: @@ -52,5 +58,7 @@ def main(): (properties['upload_user'], properties['upload_ip']), properties=properties, ensure_unique_name=True) + print("Done") + if __name__ == "__main__": main() diff --git a/bh20sequploader/qc_metadata.py b/bh20sequploader/qc_metadata.py index 78b31b2..ebe4dfc 100644 --- a/bh20sequploader/qc_metadata.py +++ b/bh20sequploader/qc_metadata.py @@ -1,6 +1,7 @@ import schema_salad.schema import logging import pkg_resources +import logging def qc_metadata(metadatafile): schema_resource = pkg_resources.resource_stream(__name__, "bh20seq-schema.yml") @@ -17,5 +18,6 @@ def qc_metadata(metadatafile): try: doc, metadata = schema_salad.schema.load_and_validate(document_loader, avsc_names, metadatafile, True) return True - except: - return False + except Exception as e: + logging.warn(e) + return False diff --git a/doc/DEVELOPMENT.md b/doc/DEVELOPMENT.md new file mode 100644 index 0000000..98d8de4 --- /dev/null +++ b/doc/DEVELOPMENT.md @@ -0,0 +1,7 @@ +# Development + +## Upload resume + +When data files get large we may want to implement resume, +like put does. See +[/sdk/python/arvados/commands/put.py](https://dev.arvados.org/projects/arvados/repository/revisions/master/entry/sdk/python/arvados/commands/put.py) diff --git a/doc/INSTALL.md b/doc/INSTALL.md new file mode 100644 index 0000000..c5c486c --- /dev/null +++ b/doc/INSTALL.md @@ -0,0 +1,31 @@ +# INSTALLATION + +Other options for running this tool. + +## GNU Guix + +Another way to install this tool is inside a [GNU Guix Environment](https://guix.gnu.org/manual/en/html_node/Invoking-guix-environment.html), which can handle installing dependencies for you even when you don't have root access on an Ubuntu system. + +1. **Set up and enter a container with the necessary dependencies.** After installing Guix as `~/opt/guix/bin/guix`, run: + +```sh +~/opt/guix/bin/guix environment -C guix --ad-hoc git python openssl python-pycurl nss-certs +``` + +2. **Install the tool.** From there you can follow the [user installation instructions](#installation-with-pip3---user). In brief: + +```sh +pip3 install --user schema-salad arvados-python-client +``` + +Pip installed the following modules + +``` +arvados-python-client-2.0.1 ciso8601-2.1.3 future-0.18.2 google-api-python-client-1.6.7 httplib2-0.17.1 oauth2client-4.1.3 pyasn1-0.4.8 pyasn1-modules-0.2.8 rsa-4.0 ruamel.yaml-0.15.77 six-1.14.0 uritemplate-3.0.1 ws4py-0.5.1 +``` + +3. Run the tool directly with + +```sh +~/opt/guix/bin/guix environment guix --ad-hoc git python openssl python-pycurl nss-certs -- python3 bh20sequploader/main.py +``` diff --git a/example/metadata.yaml b/example/metadata.yaml index 587d0be..41ff93e 100644 --- a/example/metadata.yaml +++ b/example/metadata.yaml @@ -1,17 +1,38 @@ -sampleInformation: - location: "USA" - host : "Homo Sapiens" - sequenceTechnology: "Sanger" - assemblyMethod: "CLC Genomics" +host: + host_id: XX1 + host_species: string + host_common_name: string + host_sex: string + host_age: 20 + host_age_unit: string + host_health_status: string + host_treatment: string + additional_host_information: string -InstituteInformation: - OriginatingLab: "Erik's kitchen" - SubmittingLab: "National Institute for Viral Disease Control and Prevention, China CDC" +sample: + collector_name: XXX + collecting_institution: XXX + specimen_source: XXX + collection_date: XXX + collection_location: XXX + sample_storage_conditions: XXX + additional_collection_information: XXX -SubmitterInformation: - Submitter: "National Institute for Viral Disease Control and Prevention, China CDC" - submissionDate: "04-04-2020" +virus: + virus_species: XX + virus_strain: XX -VirusDetail: - VirusName: "hCoV-19/USA/identifer/2020" - AccessionId: "EPI_ISL_Random" +technology: + sample_sequencing_technology: XX + sequence_assembly_method: XX + sequencing_coverage: 70x + +submitter: + submitter_name: tester + submitter_address: testerAdd + originating_lab: testLab + lab_address: labAdd + provider_sample_id: string + submitter_sample_id: string + authors: testAuthor + submitter_id: X12 diff --git a/example/minimal_example.yaml b/example/minimal_example.yaml new file mode 100644 index 0000000..201b080 --- /dev/null +++ b/example/minimal_example.yaml @@ -0,0 +1,14 @@ +host: + host_id: XX + host_species: string + +sample: + collector_name: XXX + collecting_institution: XXX + +technology: + sample_sequencing_technology: XX + +submitter: + submitter_name: tester + originating_lab: testLab \ No newline at end of file diff --git a/paper/paper.bib b/paper/paper.bib new file mode 100644 index 0000000..e69de29 --- /dev/null +++ b/paper/paper.bib diff --git a/paper/paper.md b/paper/paper.md new file mode 100644 index 0000000..caa9903 --- /dev/null +++ b/paper/paper.md @@ -0,0 +1,110 @@ +--- +title: 'Public Sequence Resource for COVID-19' +tags: + - Sequencing + - COVID +authors: + - name: Pjotr Prins + orcid: 0000-0002-8021-9162 + affiliation: 1 + - name: Peter Amstutz + orcid: 0000 + affiliation: 2 + - name: Tazro Ohta + orcid: 0000 + affiliation: 3 + - name: Thomas Liener + orcid: 0000 + affiliation: 4 + - name: Erik Garrison + orcid: 0000 + affiliation: 5 + - name: Michael Crusoe + orcid: 0000 + affiliation: 6 + - name: Rutger Vos + orcid: 0000 + affiliation: 7 + - Michael Heuer + orcid: 0000 + affiliation: 8 + +affiliations: + - name: Department of Genetics, Genomics and Informatics, The University of Tennessee Health Science Center, Memphis, TN, USA. + index: 1 + - name: Curii, Boston, USA + index: 2 +date: 11 April 2020 +bibliography: paper.bib +--- + +<!-- + +The paper.md, bibtex and figure file can be found in this repo: + + https://github.com/arvados/bh20-seq-resource + +To modify, please clone the repo. You can generate PDF of the paper by +pasting above link (or yours) with + + https://github.com/biohackrxiv/bhxiv-gen-pdf + +--> + +# Introduction + +As part of the one week COVID-19 Biohackathion 2020, we formed a +working group on creating a public sequence resource for Corona virus. + + +<!-- + + RESULTS! + + For each section below + + State the problem you worked on + Give the state-of-the art/plan + Describe what you have done/results starting with The working group created... + Write a conclusion + Write up any future work + +--> + +## Cloud computing backend + +Peter, Pjotr, MichaelC + +## A command-line sequence uploader + +Peter, Pjotr + +## Metadata uploader + +With Thomas + +## FASTA to GFA workflow + +Michael Heuer + +## BAM to GFA workflow + +Tazro & Erik + +## Phylogeny app + +With Rutger + +## RDF app + +Jerven? + +## EBI app + +? + +# Discussion + +Future work... + +# References |