Biological Oceanography Scientific Domain

1
Biological OceanographyScientific Domain
DataSpace

• Ed DeLong MIT
• Department of Biological Engineering Department
  of Civil and Environmental Engineering

2
BIOLOGICAL OCEANOGRAPHY

• Coupling of physical biological oceanographic
  processes
• Comparative ecosystem analysis
• Biodiversity, biomass and productivity
• C-N-P cycling and energy flow
• Production, consumption of greenhouse gases
  climate
• Measurement, modeling and experiments with
  microbial communities in the sea
• Education, training and knowledge exchange

3
Scope Scale Challenges in Biological
Oceanography
Oceanagraphic sampling approaches in the context
of scales
(Genomes to Biomes)
Microbial and sampling scales, based on Dickey
(1991) and Allen (2000) Ricardo Letelier
4
ADVANCEDINSTRUMENTATION
Continuous, autonomous collection of 4D physical
, chemical and bio-optical datasets
5
Physics

• 2 Eddies
• 1 frontal system
• Sub-mesocale features?

Biology

• Higher Chla bellow cyclone
• DCM constant
• Patchy distribution of small particles
• Advection/local production of small particles in
  the Ze

6
Further specialization Marine Metagenomics

• Traditional microbiology and microbial genome
  sequencing studies rely on cultivated cultures
• Marine metagenomics DNA sequences of microbial
  assemblages from the environment
• Metagenomic data is used by scientists across
  multiple disciplines, e.g.,
• Biological engineering biotechnology
• Genomics and computation biology
• Ecology and environmental science
• Climate relationship between marine microbes
  the oceans carbon cycle, productivity,
  greenhouse gases

7
(No Transcript)
8
2ND Gen Sequencing Platforms
AB3730
454 FLX/titan.
ILLUMINA
Cost per run 50 lt12K lt5K
Bases read/run 72 Kbp 100 Mbp 500 Mbp gt2 Gbp gt 200 Gbp !!!
Bases per read 750 250 450 gt36 (gt 100 Paired end reads)
Reads per run 96 reads/run 400K reads/run 20M reads/run
per Mbp 694 120 7
AB3730 work equivalent - 100x AB3730/dy 300x AB3730/dy
Errors Diverse (cloning bias) Homopolymeric runs Diverse (base subn.)
Run time 1 hour 6.5 hours 2-14 days
9
Biological Oceanography Data Challenges

• Wide variety and heterogeneity of data types
• Oceanographic cruise data
• Oceanographic time series data
• Laboratory field experiments
• Remote sensing datasets
• Data from gliders, AUVs moorings
• Genomics, metagenomics, gene expression data
• Numerical simulations synthesis products
• Distributed data (multi-institution
  researchers)
• Need to balance PI, project public data
  accessibility
• Data visualization analysis needs
• Long term archiving requirements

10
Why do biological oceanographers need DataSpace?
11

UH?MIT?OSU?UCSC?WHOI?MBARI
12
DataSpace partners MIT-OSU Oceanographic
Science PartnersEd DeLong (MIT) Ricardo
Letelier (OSU)Library IT PartnersMacKenzie
Smith (MIT) Terry Reese (OSU)
DeLong and Letelier Co-PIs on three major
projects Center for Microbial Oceanography
Research and Education (C-MORE) Microbial
Oceanography of Oxygen Minimum Zones (MOOMZ)
Microbial diversity and activity in seasonal
hypoxic waters (MI-LOCO)
13
Existing Data Portal
http//cmore.soest.hawaii.edu/data.htm
Currently a distributed approach. Consists of
weblinks to individually managed heterogeneous
datasets.
14
Where is the data now ? (Oceanographic data)
BCO-DMO
Biological and Chemical Oceanography Data
Management Office database
http//osprey.bcodmo.org/index.cfm
15
Where is the data now? (Genomic/metagenomic)
In-house Databases

Public Databases NCBI and CAMERANational
Center for Biotechnology Information Community
Cyberinfrastructure for Advanced Marine Microbial
Ecology Research and Analysis
http//www.ncbi.nlm.nih.gov/
http//camera.calit2.net/
16
Why do biological oceanographers need DataSpace ?

• Data access, storage, search not centralized
• Large heterogeneous datasets
• Complex data management/sharing requirements
• Shared multiple Institutions Investigator
• Long term requirements (2017)
• Need cross-investigator,institution,project
  search
• Currently lots of data is lost, e.g. not
  utilizable

17
Why do biological oceanographers need DataSpace ?
How many autonomous surveys, cruises, mooring
datasets, hydrocasts, deckboard experiments had
chlorophyll concentrations than X ? Of those
data, how many had light levels and oxygen
concentrations corresponding Y and Z ? Of
those data, how many have corresponding microbial
community taxonomic composition and gene content
data ? (retrieve) What is the relationship
between light, chlorophyll, oxygen and microbial
community taxonomic composition and gene content,
across all datasets ? How do taxa and gene
content relate to oxygen levels and the balance
of production and consumption ? Greenhouse (GHG)
gas levels ? Are there specific gene proxies
that predict oxygen or GHG levels ?
Note centralized data access, search and
storage will also drive the way we (sceintists)
ask our questions, collect, and annotate our
data. A collaboration between scientists, IT,
curators and database managers.
18
The DataSpace Project Biological Oceanography

• Enable new discoveries by facilitating access,
  search
• storage of large, complex heterogeneous datasets

• Provide infrastructure for digital archiving
  preservation
• at appropriate scales matching scope/complexity
  of data

• Enable more integrated intra- inter-project
  collaborations, analyses, data encoding,
  documentation, sharing, visualizing, and
  preservation

• Establish standards best practices to capture,
  express,
• encode and publish the policies related to
  archived data

19
The DataSpace Project Biological Oceanography