Approximately $1000 of mRNAseq will yield a
Multiple samples will allow you to generate gene
For the ascidian project I will show you,
1 graduate student,
The challenges of non-model
Missing or low quality genome reference.
Most extant computational tools focus on model
Assume low polymorphism (internal variation)
Assume reference genome
Assume somewhat reliable functional annotation
More significant compute infrastructure
…and cannot easily or directly be used on critters of
1. Challenges of non-model
2. Lamprey: too much data, not enough
3. Digital normalization as a coping
4. …applied to Molgulid ascidians…
5. …and back to lamprey.
6. More transcriptome challenges
7. What’s next?
Sea lamprey in the Great Lakes
medium to large
host fishes to
Li Lab / Y-W C-D
The problem of lamprey:
Diverged at base of vertebrates;
evolutionarily distant from model
Large, complicated genome (~2 GB)
Relatively little existing sequence.
We sequenced the liver genome…
Lamprey has incomplete genomic sequence
J. Smith et al., PNAS 2009
Evidence of somatic recombination;
100s of mb of sequence eliminated
from genome during development.
More recent evidence (unpub, J.
Smith et al.) suggests that this loss
is developmentally regulated,
results in changes in gene
expression (due to loss of genes!),
and is tissue specific.
Liver genome is not the entire
Lamprey tissues for which we have
embryo stages (late blastula,
gastrula, neurula, 22b, neural-
crest migration, 24c1,24c2)
metamorphosis 3 (intestine,
ovulatory female head skin
metamorphosis 4 (intestine,
preovulatory female eye
metamorphosis 5 (liver, intestine,
preovulatory female tail skin
metamorphosis 6 (intestine,
prespermiating male gill
freshwater (gill, intestine, kidney)
metamorphosis 7 (intestine,
mature adult male rope tissue
larval (gill, kidney, liver, intestine) monocytes
spermiating male gill
juvenile (intestine, liver, kidney) brain (0,3,21 dpi)
spermiating male head skin
lips spinal cord (0.3.21 dpi)
metamorphosis 1 (intestine,
kidney) spermiating male muscle
small parasite distal intestine,
kidney, proximal intestine
metamorphosis 2 (liver, intestine, salt water (gill, intestine)
It was the best of times, it was the wor
, it was the worst of times, it was the
isdom, it was the age of foolishness
mes, it was the age of wisdom, it was th
It was the best of times, it was the worst of times, it was
the age of wisdom, it was the age of foolishness
…but for lots and lots of fragments!
not reach the
Main problem (4 years ago):
We have a massive amount of data
that challenges existing computers
when we try to assemble it all
Solution: Digital normalization
(a computational version of library normalization)
Suppose you have a
dilution factor of A (10) to
B(1). To get 10x of B you
need to get 100x of A!
This 100x will consume
disk space and, because
of errors, memory.
We can discard it for
Digital normalization approach
A digital analog to cDNA library normalization, diginorm:
Is single pass: looks at each read only once;
Does not “collect” the majority of errors;
Keeps all low-coverage reads;
Smooths out coverage of sequencing.
=> Enables analyses that are otherwise completely
Evaluating diginorm – how?
Can’t assemble lamprey w/o
diginorm; are results any good &
how would we know?
Need comparative data set
Looking at the Molgula…
Putnam et al., 2008,
Nature.Modified from Swalla 2001
Challenging organisms to work on --
Only spawn ~1 month out of the year
Located off the northern coast of France
Hybrids not found outside of lab conditions
Species cannot be cultured
Wet lab techniques are not fully developed for species
Tail loss and notochord genes
a) M. oculata b) hybrid (occulta egg x oculata sperm) c) M. occulta
Notochord cells in orange Swalla, B. et al. Science, Vol 274, Issue 5290, 1205-1208 , 15 November 1996
Question: does it matter what
assembly pipeline you use? (No)
4 23 8 1
Diginorm V/O Raw V/O
Diginorm trinity Raw trinity
Numbers are putative orthologs (reciprocal
best hits) w/Ciona intestinalis, calculated for
How complete are these
Shift in differentially expressed genes
from gastrulation to neurulation
M. ocu vs. M. occ gastrula M. ocu vs. M. occ neurula
Differentially expressed during neurulation in M. ocu vs M. occ
Notochord gene expression similar to
-10 -5 0 5 10 15
Expression difference Hybrid vs Parent species
M. occulta transgenic NoTrlc
Alberto Stolfi & Lionel Christiaen
Lionel Christaen Claudia Racioppi
NYU Statione Zoologica Napoli
Enabling Molgula research…
Develop candidate genes to generate
hypotheses about gene network
Rapid development of genomic
resources => reporter constructs.
Doesn’t answer any biological questions
directly, but enables us to go looking for
things much faster!
We can (now) assemble really big data
sets, and get pretty good results.
We have lots of evidence (some
presented here :) that some assemblies
are not strongly affected by digital
(Note: normalization algorithm is now
standard part of Trinity mRNAseq
Transcriptome results - lamprey
Started with 5.1 billion reads from 50
(4 years of computational research, and
about 1 month of compute time, GO
Lamprey transcriptome basic
616,000 transcripts (!)
263,000 transcript families (!)
(This seems like a lot.)
Lamprey transcriptome basic
263,000 transcript families
Only 20436 transcript families have transcripts >
(compare with mouse: 17331 of 29769 genes
are > 1kb)
So, estimation by thumb ~ not that off, for long
Common vs rare genes
Can look at transcripts by tissue -
Too… many… samples…
Some known biology recapitulated; and… ???
Next steps with lamprey
Far more complete transcriptome than the one
generated from the genome!
(…but suffering from contamination,
oversensitivity to unprocessed transcripts, …?)
Enabling studies in –
Basal vertebrate phylogeny
Evolutionary origin of brown fat (previously thought
to be mammalian only!)
Pheromonal response in adults
Spinal cord regeneration
OK, we can deal with volume of data,
make pretty pictures, and ... Now what?
Both experimental or “real” contaminants are big pro
Pathway predictions vary
dramatically depending on data
The problem of lopsided gene characterization is
pervasive: e.g., the brain "ignorome"
"...ignorome genes do not differ from well-studied genes in terms of connectivity in coexpression
networks. Nor do they differ with respect to numbers of orthologs, paralogs, or protein domains.
The major distinguishing characteristic between these sets of genes is date of discovery, early
discovery being associated with greater research momentum—a genomic bandwagon effect."
Ref.: Pandey et al. (2014), PLoS One 11, e88889.lide courtesy Erich Schwarz
Practical implications of diginorm
Data is (essentially) free;
For some problems, analysis is now
cheaper than data gathering (i.e.
…plus, we can run most of our
approaches in the cloud (per-hour
rental compute resources).
Effort to provide standard “cheap”
assembly protocols for the cloud.
Entirely copy/paste; ~2-6 days from
raw reads to assembly,
annotations, and differential
Open, versioned, forkable, citable.
(“Don’t bother me unless it doesn’t
A few thoughts on our
Explicitly a “protocol” – explicit steps, copy-paste,
No requirement for computational expertise or
significant computational hardware.
~1-5 days to teach a bench biologist to use.
$100-150 of rental compute (“cloud computing”)…
…for $1000 data set.
Adding in quality control and internal validation
2. Data availability is important for
annotating distant sequences
Anything else Mollusc Cephalopod
Can we incentivize data sharing?
~$100-$150/transcriptome in the cloud
Offer to analyze people’s existing data for free,
IFF they open it up within a year.
• CephSeq white paper.
• “Dead Sea Scrolls & Open Marine Transcriptome
Project” blog post;
First results: Loligo
Putting other people’s sequences where my
w/Josh Rosenthal and Benton Grav
Lab members involved Collaborators
Adina Howe (w/Tiedje)
Billie Swalla (UW)
Josh Rosenthal (UPR)
Weiming Li, MSU
Morgan (MBL), Joe
USDA NIFA; NSF IOS;