Mashup -- NCBI Medical Genomics Resources plus Command Line Genomics; SRA-BLAST, MagicBLAST, etc.

1. Ben Busby, Ph.D.
Genomics Outreach Coordinator
NCBI
ben.busby@nih.gov
Genomic Variation in the Rising Era of Individual Genome Sequence

4.  Quick Description
 Case Study
 Medical Genetics Summary Resource
 Other Variation Resources
 Additional General Information
12/01/16 4

5. • Aggregates data from many sources
• Term hierarchy (UMLS, GeneReviews, GTR and
other vocabularies)
• Phenotypes using standard vocabularies
• Links to NCBI and outside resources
12/01/16 5

6. Now at www.omim.org
Still searchable at NCBI
• Compendium of human genes and
related phenotypes
• Each entry a review article
• Human curated from biomedical
literature
12/01/16 6
<ncbi>/omim/

7. • Clinical significance of sequence variations and
relationship to phenotypes
• Submitted and curated assertions
• Represents variants using HGVS and reference
sequences including RefSeqGene
12/01/16 7

8. • NIH’s international registry of available genetic tests voluntarily
provided by testing labs
• Submitted tests for Mendelian disorders (including pharmacogenetic tests)
• Provides searches by
• Disorder
• Test
• Laboratory
• Condition pages with links to other resources (Gene, GeneReviews)
12/01/16 8NCBI Public Services

9.  Genome Reference Consortium (GRC)
 dbGaP
 PheGenI (joint with NHGRI)
 GeneReviews
 Medical Genetics Summaries
 …and much, much more
12/01/16 9

10. The clinic secretary schedules this visit for you:
• Boy age 9 years, chief complaint:
needs medical clearance to play soccer
• Referred to genetics because of family history:
• paternal uncle died of a dissecting thoracic aortic
aneurysm at age 52
• paternal grandmother died in childbirth
You do some background reading to prepare for the case
12/01/16 10

11. Condition
Clinical feature
Gene
OMIM #
Search on “aortic dissection” which is a
clinical feature of several conditions
12/01/16 NCBI Public Services 11

12. 12/01/16 12

13. MedGen Clinical Feature Record
13

14. 14

15. 15

16. 16

17. 12/01/16 NCBI Public Services 17

18. 12/01/16 18

19. http://www.nlm.nih.gov/medlineplus/
19

20. 20

21. 21
http://www.ncbi.nlm.nih.gov/pubmedhealth/

22. 22

23. https://www.ncbi.nlm.nih.gov/pubmed/clinical/
23

24. • Patient does not meet revised Ghent criteria for Marfan
syndrome
– …but he is young and could develop diagnostic features
later in life (too young for accurate clinical diagnosis)
12/01/16 24

25. • Leading diagnosis is Marfan syndrome
– Could follow patient over time but he wants medical
release to play soccer. You are concerned about:
• Possibility of EDS IV with risk of fatal vascular rupture
• Missing the potentially severe vascular manifestations of LDS
• Familial thoracic aortic aneurysm conditions
• You decide to find a gene panel test which includes
genes for all these conditions
12/01/16 25

26. 12/01/16 26

27. 12/01/16 27

28. • Autocomplete dictionary → Specific record
• Search button → List of records that match your query
12/01/16 28

29. 29

30. 30

31. 31

32. 32

33. 33

34. 34

35.  Used MedGen to research the condition
 Used GTR to find tests
 Received lab report:
 FBN1:c.4786C>T
 Where to find information about this variant in
the fibrillin gene?
12/01/16 NCBI Public Services 35

36. 36

37. Variant
Phenotype
Submitter
FBN1:c.4786C>T
Marfan syndrome
Lab A
SCV000000010
FBN1:c.4786C>T
Marfan syndrome
Lab B
SCV000000020
Variant
Phenotype
FBN1:c.4786C>T
Marfan syndrome
RCV000000050
FBN1:c.4786C>T
Loeys Dietz syndrome
RCV000000050
FBN1:c.4786C>TVariant
37

38. 38
Practice guideline
Reviewed by expert panel
Multiple interpretations with assertion criteria
that agree
One interpretation with assertion criteria
OR multiple interpretations with assertion
criteria but conflicting
No interpretations with assertion criteria
OR no interpretation provided
http://www.ncbi.nlm.nih.gov/clinvar/docs/assertion_criteria/

39. FBN1
NM_000138.4:c.4786C>T
c.4786C>T
Arg1596Ter
R1596*
conditions
39

40. FBN1
12/01/16 NCBI Public Services 40

41. c.4786C>T
41

42. 42

43. Scroll down for evidence
43

44.  Interactively on the web; updated weekly
 Monthly full releases
 Comprehensive XML extraction
 VCF files
 Tab-delimited summary files for genes, variants
 E-utilities as web service or via command line
 Annotation on graphic sequence displays
 Variation Viewer - www.ncbi.nlm.nih.gov/variation/view/
 Variation Reporter
www.ncbi.nlm.nih.gov/variation/tools/reporter
44

45.  Diagnostic criteria are met for Marfan syndrome in
this patient
 Applied the revised Ghent nosology for diagnosing
Marfan syndrome
 Found a pathogenic variation in FBN1
45

46.  What are the guidelines for sports participation?
Address the primary reason for referral
Can he play soccer?
http://www.ncbi.nlm.nih.gov/pubmed/15184297
46
Recommendations for
this case study

47. 12/01/16 NCBI Public Services 47
‡Assumes no or only mild aortic
dilatation
*Recreational sports are categorized
with regard to high, moderate, and low
levels of exercise and graded on a
relative scale (from 0 to 5) for eligibility
with
0 to 1 indicating generally not advised or
strongly discouraged;
4 to 5 indicating probably permitted;
and 2 to 3 indicating intermediate and
to be assessed clinically on an
individual basis.
In practical terms, this means
cardiovascular evaluation for structural
defects and arrhythmias, possible
permission to play soccer if normal,
and monitoring over time

48. 48

49. Concise, structured reviews about genetic variants
and drug responses
• Includes genetic testing strategy and dosing
recommendations
• Expert-reviewed
• Regularly updated
• Free to access
• Integrated with GTR and MedGen
http://www.ncbi.nlm.nih.gov/books/NBK109194/
49

50. 50

51. 12/01/16 NCBI Public Services 51

52. 52

53.  Quick Description
 Case Study
 Medical Genetics Summary Resource
 Other Variation Resources
 Additional General Information
12/01/16 53NCBI Public Services

54. 12/01/16 NCBI Public Services 54
How does NCBI calculate clinical significance?
www.ncbi.nlm.nih.gov/variation
Which tool do I use for…?

55.  dbSNP
 Submitted (ss) and reference (rs)
 dbVar
 Structural variants (SV)
 dbGaP
 genome-wide association studies, molecular diagnostic
assays
 Controlled access to individual level data
55

56.  Quick Description
 Case Study
 Medical Genetics Summary Resource
 Other Variation Resources
 Additional General Information
56

57. https://ghr.nlm.nih.gov/primer/precisionmedicine/definition
12/01/16 NCBI Public Services 57
https://www.whitehouse.gov/precision-medicine
https://www.nih.gov/research-training/allofus-research-program

58.  Factsheet_ClinVar.pdf
 Factsheet_GTR.pdf
 Factsheet_MedGen.pdf
58
<ftp>/pub/factsheets/

59.  Factsheet_1000genomes_browser.pdf
 Factsheet_PheGenI.pdf
 Factsheet_SNP.pdf
 Factsheet_SNP_Transition_2_GRCh38.pdf
 Factsheet_Variation_Reporter.pdf
 Factsheet_Variation_Resources.pdf
 Factsheet_Variation_Viewer.pdf
 Factsheet_dbGaP.pdf
 Factsheet_dbVar.pdf
 HowTo_Finding_SNP_by_BLAST.pdf
59
<ftp>/pub/factsheets/

60.  Webinar: MedGen, GTR, and ClinVar
 Webinar: NCBI Resources and Variant Interpretation Tools
for the Clinical Community
 Webinar: NCBI Human Variation and Medical Genetics
Resources
 Search ClinVar with Ease
 Tutorials: dbGaP
 Tutorials: Genetic Testing Registry (GTR)
 Using NCBI Data with Tools that Predict the Functional
Impact of Genomic Variants
 Explore Gene Pages at NCBI: Variation & Expression
 The Variation Viewer
60

61. Graphic Credit:
Spencer Martin, UBC

62. © Martine Zilversmit 2013

63. http://1.usa.gov/1J1xmYs
NCBI NGS Online Workshop – Available on the
NCBI YouTube Channel!

64. My View of Data Transfer Principles
• Metadata Search
• Rapid NoSQL (for now)
• Integration
• Non-ambiguous identifiers
• Transferring Small amounts of Data
• Data still gets transferred in the cloud
• Underlying structure
• Finding specific data from validated formats
• Democratization of Data
• Rapid comparison by domain experts
• Reporting
• Metrics to report data upload and [unique IP] download of datasets
• Post-publication User Review
• The NCBI LinkOut Mechanism as a test suite

79. sam-dump.2.6.3 --aligned-region 17:41243452-41277500
SRR925743 > BRCA1.sam

80. (use screen or &)

86. 14,201
53,216
139,311
374,464
485,727
566,181
660,665
876,849
1,002,935
2007 2008 2009 2010 2011 2012 2013 2014 2015
Subjects

117. E-Utilities (Eutils)
Video available at:
http://www.ncbi.nlm.nih.gov/education/webinars/

118. 118
E-Utilities (Eutils)

119. 119
Introducing… Entrez Direct
The E-utilities on the UNIX
command line
esearch –db gene –query “foxp2[gene]
AND human[orgn]” |
elink –target protein –name
gene_protein_refseq |
efetch –format fasta
ftp.ncbi.nlm.nih.gov/entrez/entrezdirect/

120. 120
Edirect Cookbook

121. 121
Moving from FTP-scraping
cron jobs to on-demand APIs

122. 122
Edirect Cookbook
https://github.com/NCBI-Hackathons/EDirect_EUtils_API_Cookbook
https://ncbi-hackathons.github.io/

123. 123
New APIs!

124. 124
Generating apps that work with
our APIs and Data Structures,
and Improve Metadata:
NCBI Hackathons!

132. Combined score is
the average of SVs,
mappability, GC..
NCBI region list
Encode blacklist

133. In Twitter
@NCBI
@DCGenomics

134. NCBI Genomics Hackathon March 20-22 NIH
Campus, Bethesda
BioFrontiers Institute Hackathon May 22-24,
Boulder, CO
NYGC Hackathon June 2017

135. u Basespace (Illumina)
v Independent Consultants
w mothur (for metagenomics)
x CyVerse (iPlant)
y NCBI Submission Portal
• If you have another submission platform, or offer this as a service, please send
us an email at webinars@ncbi.nlm.nih.gov, and we will include it in the FTP release notes

136. Workflow | Storage | Analysis | Sharing
>70
Apps
>45,000
Users
>240,000
Runs
>6,000
Instruments
BaseSpace Labs Apps
Illumina Core Applications
Third-Party Applications

137. Easily Import Samples into BaseSpace with the SRA Import App
1. Launch the SRA Import
App
2. Input Accession
Numbers
3. View Downloaded
FASTQs (Samples)
Launch the SRA Import app
from the Apps page
Enter SRA Accession
number(s) and Continue
After app completes, view
the imported Samples in
your BaseSpace project

138. Submit Your Data Using the SRA Submission App
1. Register a BioSample and
BioProject with NCBI
2. Launch the SRA Submission
App
3. Enter your submission
information
Launch the SRA Submission
app from the Apps page
Fill out the input form with
your submission and sample
details
4. When the app completes, receive email confirmations from NCBI
Receive emails from NCBI upon data receipt and once the data is available in SRA

139. www.illumina.com/BaseSpaceApps
www.basespace.com
www.blog.basespace.illumina.com
BaseSpace Information View Demo information
Contact Information
Log in to your BaseSpace account and view an
example SRA Import app session:
https://basespace.illumina.com/s/Zls7s5ZF
O8ns
Jay Patel, Assoc. Product Manager, BaseSpace Applications
jpatel@illumina.com

140.  Originally published in 2009
(doi:10.1128/AEM.01541-09)
 Most cited tool for analyzing 16S
rRNA gene sequences
 3,410 citations (WoS: 1/8/2016)
 Working on 37th release
 Overview
 100% open source, GPL v3
 OS independent
 Command line interface
 Written in C/C++
http://www.mothur.org

141.  Deposition of 16S rRNA gene sequences to the SRA has
been a major problem
 Worked with SRA staff to make a customized portal to
simplify submission of PCR-generated 16S rRNA gene
sequences
 Command enforces co-submission of sample and
processing metadata
 Originally released in March 2015. So far 86 submissions,
61 studies, 6367 runs, 116 Gbp total
http://www.mothur.org/wiki/make.sra

142. 1. Provide the necessary MIMARKS* metadata data about
samples with get.mimarkspackage
2. Create a project file describing user and their project
using supplied template file
3. Parse MIMARKS, project file, and sff or fastq files to
generate an xml file for submission using make.sra
4. Email the SRA to let them know about submission using
mothur created files and await further instructions
* http://www.ncbi.nlm.nih.gov/pmc/articles/PMC3367316/
http://www.mothur.org/wiki/Creating_a_new_submission

143.  MIMARKS: minimum information about a marker gene
sequence (doi:10.1038/nbt.1823)
 Command supplies user with a blank text file with sample
names and necessary metadata for their environment.
User fills in details.
 For each environmental package and environment there
is a wiki page with required and optional parameters and
allowed values. Can also be extended to include
additional metadata
http://www.mothur.org/wiki/Get.mimarkspackage
http://www.mothur.org/wiki/Human_gut

144. USERNAME [UserName]
Last [LastName]
First [FirstName]
EMAIL [Email@mail.com]
CENTER [University or Center Name]
TYPE institute
WEBSITE [www.Website.org]
ProjectName [ProjectName]
ProjectTitle [Project Title]
Description [Project Description]
Grant id=[GrantID], agency=[GrantAgency],
title=[GrantTitle]
User completes information in brackets
http://www.mothur.org/wiki/Project_File

145. CyVerse-Enabled SRA Submission Pipeline
• 10-year, $100 million, NSF-funded mandate to support all of Life
Science (formerly The iPlant Collaborative)
• “Transforming science through data-driven discovery.”
• People + Cyberinfrastructure, empowering researchers, and
fostering interoperability
www.cyverse.org jdebarry@cyverse.org

146.  GUI connection to CyVerse Data Store
 Built on iRODS, supports ~30k users and >1.25 PB of data
 A platform that can run almost any bioinformatics application
 Seamlessly integrated with data and HPC resources
 Point and click tools for data and metadata management
Easy, fast, secure data transfer, management, and analysis
CyVerse Discovery Environment
Submit to SRA via Discovery Environment
• Submission package = sequence data and metadata XML
• Checksums, etc. are automatically created
• Create or update BioProject during SRA submission
• Create BioSample(s) during SRA submission

147. Submission tutorial (https://goo.gl/LMe5kQ)
with video instructions and example
submission package in the CyVerse wiki
1. Efficient command line and point and
click tools available to transfer data to
the Discovery Environment
 CyVerse supports
 Apps available to compress data if needed
2. Create submission package folder with
dedicated tool
 Single BioProject folder contains one or
more BioSample folders, each with one or
more Library folders
 Drag and drop sequence files to organize
submission package
4) Submit data
and metadata to
SRA
2) Create
submission
package
3) Enter and
save metadata
5) Submission
notification
from SRA
1) Upload data
to Discovery
Environment
6) Error
Correction
(if needed)

148. 3. BioProject, BioSample, Sequencing Library metadata entered via templates
 Choose from available templates to create/update BioProject, appropriate BioSample type
 Metadata term guide in the Discovery Environment defines metadata fields
 Metadata applied to each folder, copy metadata to folders to limit entry for large submissions
 After metadata entry, save single file of metadata for submission package
4) Submit data
and metadata to
SRA
2) Create
submission
package
3) Enter and
save metadata
5) Submission
notification
from SRA
1) Upload data
to Discovery
Environment
6) Error
Correction
(if needed)

149. 4. Discovery Environment App requires only top-level BioProject folder and file of
saved metadata as input
 App creates XML metadata file and submits to SRA via Aspera Connect
5. SRA uses contact email in submission metadata to transfer ownership to
submitter’s NCBI account
 Notification emails for successful submission or necessary error corrections sent to submitter
6. Discovery Environment App can retrieve error report from SRA
 To correct, edit metadata and submission package and resubmit from Discovery Environment
4) Submit data
and metadata to
SRA
2) Create
submission
package
3) Enter and
save metadata
5) Submission
notification
from SRA
1) Upload data
to Discovery
Environment
6) Error
Correction
(if needed)

153. Pro tip:
Before starting your
submission;
<google> for
“biosample template”
and collect your
metadata

155. Biosample

156. The Future

157. The Future (in my
opinion)

158. The Future (in my
opinion)…
Is already here

162. Integration into a Larger Data
Discovery Framework
BD2K - bioCADDIE

163. Integration into a Larger Data
Discovery Framework

164. Integration into a Larger Data
Discovery Framework
Example: GOLD (JGI)

165. 165
New APIs!

166. 166
Edirect Cookbook
https://github.com/NCBI-Hackathons/EDirect_EUtils_API_Cookbook
https://ncbi-hackathons.github.io/

169. My View of Data Transfer Principles
• Metadata Search
• Rapid NoSQL (for now)
• Integration
• Non-ambiguous identifiers
• Transferring Small amounts of Data
• Data still gets transferred in the cloud
• Underlying structure
• Finding specific data from validated formats
• Democratization of Data
• Rapid comparison by domain experts
• Reporting
• Metrics to report data upload and [unique IP] download of datasets
• Post-publication User Review