A
human
commons
cell
atlas
reveals
cell
type
specificity
for
OAS1
isoforms
Ángel
Galvez-Merchán
1*
,
A.
Sina
Booeshaghi
2*
&
Lior
Pachter
3,4+
1.
Cellarity,
Somerville,
MA,
USA
2.
Department
of
Bioengineering,
University
of
California
Berkeley,
Berkeley,
CA,
USA
3.
Division
of
Biology
and
Biological
Engineering,
California
Institute
of
Technology,
Pasadena,
CA,
USA
4.
Department
of
Computing
&
Mathematical
Sciences,
California
Institute
of
Technology,
Pasadena,
CA,
USA
*These
authors
contributed
equally
+To
whom
correspondence
should
be
addressed.
Supplementary
Figure
1:
Figure
S1:
Scatterplot
showing
the
relationship
between
the
number
of
reads
in
the
raw
FASTQ
and
the
pre-processing
runtime
in
seconds
for
each
sample
of
the
human
CCA.
The
runtime
refers
to
the
‘
kb
count
’
command,
which
calls
the
following
commands
in
the
following
order:
‘
kallisto
bus
’,
‘
bustools
inspect
’,
‘
bustools
sort
’,
‘
bustools
inspect
’,
‘
bustools
inspect
’,
’
bustools
correct
’,
‘
bustools
inspect
’,
’
bustools
sort
’,
‘
bustools
inspect
’,
‘
bustools
count
’,
‘
bustools
whitelist
’,
‘
bustools
correct
’,
‘
bustools
inspect
’,
‘
bustools
sort
’,
‘
bustools
inspect
’,
‘
bustools
count
’.
Supplementary
Figure
2:
Figure
S2
:
Normalised
expression
of
OAS1
isoforms
across
all
cells
in
the
human
CCA.
The
most
widely
expressed
isoforms
(p42,
p44b,
p56
and
p48)
are
highlighted
in
blue.
Supplementary
Figure
3:
Figure
S3
:
The
BAM
file
for
one
of
the
testis
observations
(GSM3302525)
was
downloaded
from
GEO
and
visualized
using
the
Integrative
Genomics
Viewer
(IGV).
Over
500
reads
mapping
to
exon
8
and
the
p44b-specific
intron
were
detected.
Supplementary
Figure
4:
Figure
S4
:
Isoform
expression
of
the
4
main
OAS1
isoforms
in
Round
Spermatids
split
by
paper,
showing
that
the
expression
specificity
of
the
isoform
p44b
reproduces
across
independent
studies
and
datasets.
Supplementary
Figure
5:
Figure
S5
:
The
raw
TCC
matrix
of
the
testis
sample
GSM2928378
was
filtered
to
counts
that
mapped
to
a
single
equivalence
class,
which
were
averaged
across
all
cells.
A
isoform-to-gene
ratio
was
calculated
by
dividing
the
result
of
each
isoform
by
the
averaged
raw
counts
of
the
corresponding
gene
expression.
For
each
isoform,
this
ratio
was
plotted
against
the
averaged
raw
counts
of
the
corresponding
gene.
Supplementary
Figure
6:
Figure
S6
:
Genes
that
experience
cell-type-specific
isoform
switching
were
manually
selected,
and
the
mean
isoform
expression
for
cells
belonging
or
not
to
the
cell-type
was
plotted.
Supplementary
Figure
7:
Figure
S7
:
The
CONCORDEX
ratio
(citation)
of
each
sample
was
calculated
using
the
assignments
derived
from
‘mx
assign’.
The
gene
markers
for
each
organ
were
validated
by
calculating
the
average
CONCORDEX
ratio
per
organ.
Organs
with
high
CONCORDEX
ratio
are
shown
in
the
plot.