across the eight alternative models. Genes and
transcripts were required to have a posterior
probability > 0.5 and a fold change in expression > 1 to be declared cell type–specific. To compare
cell type–specific gene expression estimates between our RNA-seq data and publicly available
microarray data sets, we retrieved probe annotations for the Illumina (16) and Affymetrix (14)
platforms from Ensembl v70.
Splice junction analysis
Identification of splice junctions for each sample was based on the alignment of the trimmed
reads to the human genome (GRCh37) with
three different aligners: GSNAP (59), STAR (60),
and GEM (61). Splice junctions were considered
for further analysis if supported by all three
aligners and by at least 10 reads in at least two
samples, where reads covered a minimum of
10 bp at both ends of the splice junction. We
defined a splice junction as unannotated if not
present in Ensembl v70. These were further
compared to the EST/mRNA data from UCSC
and the Illumina BodyMap 2.0 data set (35) to
identify novel splice junctions.
Splice site probability scores were extracted
from the GSNAP output. PhastCons conservation scores were used to plot evolutionary conservation in the 100 bp surrounding each splice
junction. Coding potential was estimated by
summing the number of possible stop codons in
exonic and intronic regions, in all reading frames
and in 100 bp flanking the unannotated splice
site. Shannon’s entropy (37) was calculated on the
basis of the read coverage of the splice junctions.
DSU was identified with a beta-binomial model.
The characterization of the protein domains in
cassette exons with DSU was performed with
InterProscan 5 (62) to search for domains predicted by Pfam.
Validation of transcript isoform
expression and splicing events
To validate the quantification of transcript levels determined by analysis of the RNA-seq data
with MMSEQ, we performed RT-PCR assays
with 40 transcript-specific assays and 5 positive
control assays in multiple cell subsets. Quantita-tion of each transcript was performed in multiple progenitor cell subsets with the BioMark HD
system (Fluidigm, San Francisco, California). After requiring call quality scores >0.9 (Fluidigm
Real-Time PCR Analysis software, www.fluidigm.
com/ software.html), 36 transcripts were analyzed.
For each probe, DDCq values were calculated with
the B2M transcript as control and the average
DDCq for MKs. Linear regression was then performed between DDCq values and the corresponding MMSEQ estimates relative to mean MK.
To validate progenitor-specific novel splice junctions and exon-skipping events, we designed PCR
primers to amplify 30 junctions identified by RNA-seq. PCR was performed on pools of the RNA-seq
libraries. PCR products were run on agarose gel
and imaged, and densitometry was performed.
PacBio libraries were generated (Pacific Biosci-
ences, Menlo Park, California) from cDNA obtained
by reverse transcription of 10 ng of MK_3 total
RNA and sequenced in five SMRT cells on the
PacBio RSII. SMRT Pipe and ICE were used to
filter reads to generate consensus sequence clusters
that were mapped to the reference human ge-
nome (GRCh37) using GMAP.
Enrichment analysis of RNA-binding
motifs around cassette exons
Motif enrichment analysis of 102 RNA-binding
motifs (41) was performed on DSU cassette exons
(FDR < 0.05 and usage proportion change > 0.05)
over three genomic regions: upstream intronic
(300 bp), exonic, and downstream intronic (300 bp).
Enrichment and depletion of RNA-binding motifs was determined with cumulative hypergeometric testing, and P values were corrected for
Cloning, shRNA, lentivirus production,
TRC shRNA lentivirus targeting NFIB and NFIC
and a nonsilencing control were purchased from
Thermo Scientific Open Biosystems (Little Chalfont,
UK). NFIB full length and NFIC cDNA were
cloned into p WPI TAP-tagged vector. Packaging
was performed in 293T cells, and viral stocks were
titrated and quantified using qPCR and, for p WPI,
using qPCR and green fluorescent protein FACS.
CD34+ cells were purified from NHS Blood and
Transplant apheresis filters, as above (Miltenyi), and
then infected with lentiviral particles in the presence of polybrene, in medium supplemented with
thrombopoietin and IL-1b. On day 2, medium was
replaced and cells were cultured toward MKs. At
day 10, MKs were counted and assessed by morphology and flow cytometry for maturation.
and Western blots
To detect protein-protein interactions, NFI proteins were expressed by cotransfection in 293T
cells and immunoprecipitated with an anti-Flag
antibody. Western blots were probed with NFIB,
NFIC, and b-tubulin.
River plots were generated with the ggplot2 R
package (version 0.9.3.1), which was obtained from
www.bioconductor.org/). Heat maps
were generated with both the gplots (version 2.13.0)
and pheatmap (version 0.7.7) R packages, which
were both obtained from CRAN (http://cran.
r-project.org/). For sequence logos of the splice site
motifs, we used seqLogo R package (version 1.30.0)
available from Bioconductor. The IGV genome
browser (version 2.3.34) was used for visualization
www.broadinstitute.org/igv/). For further details,
please refer to the supplementary materials.
REFERENCES AND NOTES
1. S. H. Orkin, L. I. Zon, Hematopoiesis: An evolving paradigm for
stem cell biology. Cell 132, 631–644 (2008). doi: 10.1016/
j.cell.2008.01.025; pmid: 18295580
2. J. Hoggatt, J. M. Speth, L. M. Pelus, Concise review: Sowing
the seeds of a fruitful harvest: Hematopoietic stem cell
mobilization. Stem Cells 31, 2599–2606 (2013). doi: 10.1002/
stem. 1574; pmid: 24123398
3. I. I. Slukvin, Hematopoietic specification from human
pluripotent stem cells: Current advances and challenges
toward de novo generation of hematopoietic stem cells. Blood
122, 4035–4046 (2013). doi: 10.1182/blood-2013-07-474825;
4. M. P. Lambert, S. K. Sullivan, R. Fuentes, D. L. French,
M. Poncz, Challenges and promises for the development of
donor-independent platelet transfusions. Blood 121, 3319–3324
(2013). doi: 10.1182/blood-2012-09455428; pmid: 23321255
5. B. Giebel, I. Bruns, Self-renewal versus differentiation in
hematopoietic stem and progenitor cells: A focus on
asymmetric cell divisions. Curr. Stem Cell Res. Ther.
3, 9–16 (2008). doi: 10.2174/157488808783489444;
6. R. Majeti, C. Y. Park, I. L. Weissman, Identification of a
hierarchy of multipotent hematopoietic progenitors in human
cord blood. Cell Stem Cell 1, 635–645 (2007). doi: 10.1016/
j.stem.2007.10.001; pmid: 18371405
7. C. M. Baum, I. L. Weissman, A. S. Tsukamoto, A. M. Buckle,
B. Peault, Isolation of a candidate human hematopoietic
stem-cell population. Proc. Natl. Acad. Sci. U.S.A. 89, 2804–2808
(1992). doi: 10.1073/pnas.89.7.2804; pmid: 1372992
8. S. H. Orkin, Diversification of haematopoietic stem cells
to specific lineages. Nat. Rev. Genet. 1, 57–64 (2000).
doi: 10.1038/35049577; pmid: 11262875
9. S. Doulatov, F. Notta, E. Laurenti, J. E. Dick, Hematopoiesis:
A human perspective. Cell Stem Cell 10, 120–136 (2012).
doi: 10.1016/ j.stem.2012.01.006; pmid: 22305562
10. H. Iwasaki et al., The order of expression of transcription
factors directs hierarchical specification of hematopoietic
lineages. Genes Dev. 20, 3010–3021 (2006). doi: 10.1101/
gad.1493506; pmid: 17079688
11. J. Mestas, C. C. Hughes, Of mice and not men: Differences
between mouse and human immunology. J. Immunol. 172,
2731–2738 (2004). doi: 10.4049/jimmunol.172.5.2731;
12. D. G. Tenen, R. Hromas, J. D. Licht, D. E. Zhang, Transcription
factors, normal myeloid development, and leukemia. Blood 90,
489–519 (1997). pmid: 9226149
13. R. E. Dickinson et al., Exome sequencing identifies
GATA-2 mutation as the cause of dendritic cell, monocyte,
B and NK lymphoid deficiency. Blood 118, 2656–2658
(2011). doi: 10.1182/blood-2011-06-360313; pmid: 21765025
14. N. Novershtern et al., Densely interconnected transcriptional
circuits control cell states in human hematopoiesis. Cell
144, 296–309 (2011). doi: 10.1016/j.cell.2011.01.004;
15. N. A. Watkins et al., A HaemAtlas: Characterizing gene
expression in differentiated human blood cells. Blood 113,
e1–e9 (2009). doi: 10.1182/blood-2008-06-162958;
16. E. Laurenti et al., The transcriptional architecture of early
human hematopoiesis identifies multilevel control of lymphoid
commitment. Nat. Immunol. 14, 756–763 (2013). doi: 10.1038/
ni.2615; pmid: 23708252
17. T. W. Nilsen, B. R. Graveley, Expansion of the eukaryotic
proteome by alternative splicing. Nature 463, 457–463 (2010).
doi: 10.1038/nature08909; pmid: 20110989
18. E. T. Wang et al., Alternative isoform regulation in human
tissue transcriptomes. Nature 456, 470–476 (2008).
doi: 10.1038/nature07509; pmid: 18978772
19. Q. Pan, O. Shai, L. J. Lee, B. J. Frey, B. J. Blencowe, Deep
surveying of alternative splicing complexity in the human
transcriptome by high-throughput sequencing. Nat. Genet. 40,
1413–1415 (2008). doi: 10.1038/ng.259; pmid: 18978789
20. N. L. Barbosa-Morais et al., The evolutionary landscape of
alternative splicing in vertebrate species. Science 338,
1587–1593 (2012). doi: 10.1126/science.1230612;
21. J. Merkin, C. Russell, P. Chen, C. B. Burge, Evolutionary
dynamics of gene and isoform regulation in mammalian
tissues. Science 338, 1593–1599 (2012). doi: 10.1126/
science.1228186; pmid: 23258891
22. R. K. Singh, T. A. Cooper, Pre-mRNA splicing in disease and
therapeutics. Trends Mol. Med. 18, 472–482 (2012).
doi: 10.1016/ j.molmed.2012.06.006; pmid: 22819011
23. K. Yoshida et al., Frequent pathway mutations of splicing
machinery in myelodysplasia. Nature 478, 64–69 (2011).
doi: 10.1038/nature10496; pmid: 21909114
24. M. Cazzola, M. G. Della Porta, L. Malcovati, The genetic basis
of myelodysplasia and its clinical relevance. Blood 122,
4021–4034 (2013). doi: 10.1182/blood-2013-09-381665;