def plot_dendrogram(obj_arr,
n_by_mad=None,
qn_method=None,
eps=0.01,
min_val=1,
n_above_min=3,
vertical=False,
figsize=(7, 8),
**kwargs):
if len(obj_arr) > 1:
the_obj = loader.MultipleBatchLoader(obj_arr)
else:
the_obj = obj_arr[0]
the_obj = filter_loader(the_obj, min_val=min_val, n_above_min=n_above_min)
dat = np.log2(the_obj.data + eps)
if qn_method is not None:
dat = transformations.quantile_normalisation(dat, method=qn_method)
if n_by_mad is not None:
mad = transformations.median_absolute_deviation(dat).sort_values(
ascending=False)
dat = dat.loc[mad.index[:n_by_mad]]
cc, st, leg_dict = construct_colour_array_legend_studies(the_obj.meta)
dend = clustering.dendrogram_with_colours(dat,
cc,
vertical=vertical,
legend_labels=leg_dict,
fig_kws={'figsize': figsize},
**kwargs)
return dend
# obj.filter_samples(ix)
#
# ix = obj.meta.type != 'iAPC'
# obj.filter_samples(ix)
#
# ix = ~obj.meta.index.str.contains('GBM')
# obj.filter_samples(ix)
# ix = obj.meta.index != 'H9 NPC (Encode EPIC)'
# obj.filter_samples(ix)
bdat = obj.data
mdat = process.m_from_beta(bdat)
if qn_method is not None:
mdat = transformations.quantile_normalisation(mdat, method=qn_method)
# tidy up batch IDs
obj.meta.loc[obj.meta.batch.isnull(),
'batch'] = obj.meta.loc[obj.meta.batch.isnull(), 'batch_1']
obj.meta.batch = obj.meta.batch.str.replace('2016-12-19_ucl_genomics',
'2016-12-19')
# the only batch names without letters are ours
obj.meta.loc[~obj.meta.batch.str.contains(r'[A-Z]'),
'batch'] = 'This study'
# PCA plot (by batch and cell type)
colour_subgroups = obj.meta.batch
c_sub_sorted = sorted(colour_subgroups.unique(),
key=lambda x: 'A' if x == 'This study' else x)
rna_cc_dat = filter.filter_by_cpm(rna_cc_dat,
min_cpm=min_tpm,
min_n_samples=2)
# reduce to matching probes
probes = rna_cc_dat.index.intersection(rna_ff_dat.index)
if remove_mt:
probes = probes[~probes.isin(mt_ens)]
rna_ff_dat = np.log2(rna_ff_dat.loc[probes] + eps)
rna_cc_dat = np.log2(rna_cc_dat.loc[probes] + eps)
# QN
if apply_qn:
rna_ff_dat = transformations.quantile_normalisation(rna_ff_dat)
rna_cc_dat = transformations.quantile_normalisation(rna_cc_dat)
# correlation plot
pdist = pd.DataFrame(index=rna_ff_dat.columns.sort_values(),
columns=rna_cc_dat.columns.sort_values(),
dtype=float)
for ff in pdist.index:
for cc in pdist.columns:
if dist_metric == 'pearson':
pdist.loc[ff, cc] = stats.pearsonr(rna_ff_dat[ff],
rna_cc_dat[cc])[0]
elif dist_metric == 'spearman':
pdist.loc[ff, cc] = stats.spearmanr(rna_ff_dat[ff],
rna_cc_dat[cc]).correlation
else:
if labels_included[csg]:
lbl = None
else:
lbl = csg
labels_included[csg] = True
ax.plot(xi, yi, c=colour_map[csg], label=lbl)
ax.set_xlabel('M value')
ax.set_ylabel('ECDF')
ax.set_title(ct)
common.legend_outside_axes(ax)
fig.subplots_adjust(left=0.1, bottom=0.1, right=0.8, top=0.99)
fig.savefig(os.path.join(outdir, "ecdf_batches_%s.png" % ct), dpi=200)
# normalise and repeat PCA
mdat_qn = transformations.quantile_normalisation(mdat)
fig = plt.figure(figsize=(10, 7))
ax = fig.add_subplot(111)
p_qn, ax = plot_pca(mdat_qn,
colour_subgroups,
marker_subgroups=m_subgroups,
marker_map=mmap,
ax=ax)
ax.figure.subplots_adjust(left=0.1, right=0.8)
ax.figure.savefig(os.path.join(outdir,
"pca_plot_batch_cell_type_all_qn.png"),
dpi=200)
# now try without GBM
# (a) no QN
obj1.filter_samples(ix)
dend = plot_dendrogram([obj1, ref_obj, nsc_ref_obj],
vertical=False,
figsize=(7, 14),
qn_method=quantile_norm,
n_by_mad=n_gene_by_mad)
dend['fig'].savefig(os.path.join(outdir, "cluster_ipsc_esc_fb_nsc.png"),
dpi=200)
# 1b. Heatmap from clustering result of (1a)
n_for_heatmap = 500
the_obj = loader.MultipleBatchLoader([obj1, ref_obj, nsc_ref_obj])
the_dat = np.log2(the_obj.data + eps)
if quantile_norm is not None:
the_dat = transformations.quantile_normalisation(the_dat,
method=quantile_norm)
the_mad = transformations.median_absolute_deviation(the_dat).sort_values(
ascending=False)
cc, st, leg_dict = construct_colour_array_legend_studies(the_obj.meta)
# ref line colours
for k, v in cell_line_colours.items():
cc.loc[the_obj.meta.type == k, 'Cell type'] = v
# our line colours
cc.loc[the_obj.meta.batch.str.contains('wtchg') &
(the_obj.meta.type == 'iNSC'),
'Cell type'] = cell_line_colours['iNSC (this study)']
cc.loc[the_obj.meta.batch.str.contains('wtchg') &
(the_obj.meta.type == 'iPSC'),
'Cell type'] = cell_line_colours['iPSC (this study)']
def plot_clustermap(obj,
quantile_norm,
method='average',
metric='correlation',
n_gene_by_mad=5000,
n_gene_for_heatmap=500,
fmin=0.05,
fmax=0.95,
eps=0.01,
cell_line_colours=None):
if cell_line_colours is None:
cell_line_colours = {
'FB': '#fff89e', # yellow
'GBM (this study)': '#e6e6e6', # light gray
'GBM': '#4d4d4d', # dark grey
'ESC': '#ff7777', # light red
'iPSC': '#990000', # dark red
'iPSC (this study)': '#fdc086', # orange
'NSC': '#006600', # dark green
'iNSC (this study)': '#7fc97f', # green
}
the_dat = np.log2(obj.data + eps)
if quantile_norm is not None:
the_dat = transformations.quantile_normalisation(the_dat,
method=quantile_norm)
the_mad = transformations.median_absolute_deviation(the_dat).sort_values(
ascending=False)
cc, st, leg_dict = construct_colour_array_legend_studies(obj.meta)
# linkage
lkg = hc.linkage(
the_dat.loc[the_mad.index[:n_gene_by_mad]].transpose(),
method=method,
metric=metric,
)
# ref line colours
for k, v in cell_line_colours.items():
cc.loc[obj.meta.type == k, 'Cell type'] = v
# our line colours
cc.loc[obj.meta.batch.str.contains('wtchg') & (obj.meta.type == 'iPSC'), 'Cell type'] = \
cell_line_colours['iPSC (this study)']
# get appropriate clims
the_dat = the_dat.loc[the_mad.index[:n_for_heatmap]]
the_dat_flat = np.sort(the_dat.values.flatten())
vmin = the_dat_flat[int(len(the_dat_flat) * fmin)] - 0.5
vmax = the_dat_flat[int(len(the_dat_flat) * fmax)] + 0.5
gc = clustering.plot_clustermap(
the_dat.loc[the_mad.index[:n_gene_for_heatmap]],
cmap='RdBu_r',
col_linkage=lkg,
col_colors=cc,
vmin=vmin,
vmax=vmax,
)
leg_entry = {
'class': 'patch',
'edgecolor': 'k',
'linewidth': 1.,
}
leg_dict2 = collections.OrderedDict()
leg_dict2['Cell type'] = collections.OrderedDict()
for k in sorted(cell_line_colours):
if k.replace(' (this study)', '') in obj.meta.type.unique():
leg_dict2['Cell type'][k] = dict(leg_entry)
leg_dict2['Cell type'][k].update(
{'facecolor': cell_line_colours[k]})
leg_dict2['Study'] = {}
for k, v in leg_dict['Study'].items():
leg_dict2['Study'][k] = dict(leg_entry)
leg_dict2['Study'][k].update({'facecolor': v})
common.add_custom_legend(gc.ax_heatmap,
leg_dict2,
loc_outside=True,
fontsize=14)
format_clustermap(gc)
return gc
