def main(dir):
save_dir = os.path.join(dir, "plots", "phase")
container = NDataContainer(load_on_fly=True)
container.add_axona_files_from_dir(dir)
container.setup()
spike_names = container.get_file_dict()["Spike"]
for i, ndata in enumerate(container):
name = spike_names[container._index_to_data_pos(i)[0]]
results = ndata.phase_at_spikes(should_filter=True)
positions = results["positions"]
phases = results["phases"]
good_place = results["good_place"]
directions = results["directions"]
co_ords = {}
co_ords["north"] = np.nonzero((45 <= directions) & (directions < 135))
co_ords["south"] = np.nonzero((225 <= directions)
& (directions <= 315))
if (phases.size != 0) and good_place:
for direction in "north", "south":
dim_pos = positions[1][co_ords[direction]]
directional_phases = phases[co_ords[direction]]
fig, ax = plt.subplots()
ax.scatter(dim_pos, directional_phases)
# ax.hist2d(dim_pos, directional_phases, bins=[10, 90])
parts = os.path.basename(name[0]).split(".")
end_name = (parts[0] + "_unit" + str(ndata.get_unit_no()) +
"_" + direction + ".png")
out_name = os.path.join(save_dir, end_name)
make_dir_if_not_exists(out_name)
fig.savefig(out_name)
plt.close(fig)
def main(dir):
container = NDataContainer(load_on_fly=True)
container.add_axona_files_from_dir(dir, recursive=True)
container.setup()
print(container.string_repr(True))
nca.place_cell_summary(
container, dpi=200, out_dirname="nc_spat_plots", output_format="pdf",
burst_thresh=6, isi_bin_length=1, filter_place_cells=False)
def main(dir, bin_size, bin_bound):
container = NDataContainer(load_on_fly=True)
container.add_axona_files_from_dir(dir, recursive=True)
container.setup()
print(container.string_repr(True))
event = NEvent()
last_stm_name = None
dict_list = []
for i in range(len(container)):
try:
result_dict = OrderedDict()
data_index, _ = container._index_to_data_pos(i)
stm_name = container.get_file_dict("STM")[data_index][0]
data = container[i]
# Add extra keys
spike_file = data.spike.get_filename()
spike_dir = os.path.dirname(spike_file)
spike_name = os.path.basename(spike_file)
spike_name_only, spike_ext = os.path.splitext(spike_name)
spike_ext = spike_ext[1:]
result_dict["Dir"] = spike_dir
result_dict["Name"] = spike_name_only
result_dict["Tet"] = int(spike_ext)
result_dict["Unit"] = data.spike.get_unit_no()
# Do analysis
if last_stm_name != stm_name:
event.load(stm_name, 'Axona')
last_stm_name = stm_name
graph_data = event.psth(data.spike, bins=bin_size, bound=bin_bound)
spike_count = data.get_unit_spikes_count()
result_dict["Num_Spikes"] = spike_count
# Bin renaming
for (b, v) in zip(graph_data["all_bins"][:-1], graph_data["psth"]):
result_dict[str(b)] = v
dict_list.append(result_dict)
# Do plotting
name = (spike_name_only + "_" + spike_ext + "_" +
str(result_dict["Unit"]) + ".png")
plot_name = os.path.join(dir, "psth_results", name)
make_dir_if_not_exists(plot_name)
plot_psth(graph_data, plot_name)
print("Saved psth to {}".format(plot_name))
except Exception as e:
log_exception(e, "During stimulation batch at {}".format(i))
dict_list.append(result_dict)
fname = os.path.join(dir, "psth_results", "psth.csv")
save_dicts_to_csv(fname, dict_list)
print("Saved results to {}".format(fname))
def main(dir):
container = NDataContainer(load_on_fly=True)
container.add_axona_files_from_dir(dir, recursive=True)
container.setup()
print(container.string_repr(True))
nca.place_cell_summary(container, dpi=200, out_dirname="nc_spat_plots")
def main(in_dir, tetrode):
container = NDataContainer(load_on_fly=True)
regex = ".*objects.*"
container.add_axona_files_from_dir(
in_dir, True, False, tetrode_list=[tetrode], re_filter=regex)
container.setup()
out_dir = os.path.join(in_dir, "nc_results")
make_dir_if_not_exists(out_dir)
out_dict = OrderedDict()
headers = []
base_list = ["NW", "NE", "SW", "SE"]
for ap in ["Spikes", "Rate", "Norm_Spikes", "Norm_Rate"]:
mod_list = [b + "_" + ap for b in base_list]
headers = headers + mod_list
out_dict["File"] = headers
out_vec = OrderedDict()
out_vec, out_dict = calculate_directional_stats(
container, out_vec, out_dict)
out_dict["Summary Stats Rate"] = [
"Rot_Dist", "Rot_U_Dist", "Shuf_Dist", "Shuf_U_Dist", "Shuf_UR_Dist"]
out_dict["Summary Stats Rate"] += ["Rate" + b for b in base_list]
out_dict["Summary Stats Rate"] += ["Rot Rate" + b for b in base_list]
out_dict["Summary Stats Rate"] += ["Undo Rot Rate" + b for b in base_list]
out_dict["Summary Stats Rate"] += ["Shuf Rate" + b for b in base_list]
out_dict["Summary Stats Rate"] += ["Undo Shuf Rate" + b for b in base_list]
out_dict["Summary Stats Rate"] += ["Undo ShufR Rate" + b for b in base_list]
for key, vec in out_vec.items():
res, p_vecs, pu_vecs, ur = distance_between(
vec, key=key, measure=euc_dist)
out_dict["Rate Unit " + str(key)] = np.concatenate(
[res, p_vecs[0], p_vecs[1], pu_vecs[1], p_vecs[2], pu_vecs[2], ur])
fig, ax = plt.subplots()
heat_arr = np.zeros(shape=(2, 12))
heat_arr[:, :2] = p_vecs[0].reshape(2, 2)
heat_arr[:, 2:4] = p_vecs[1].reshape(2, 2)
heat_arr[:, 4:6] = pu_vecs[1].reshape(2, 2)
heat_arr[:, 6:8] = p_vecs[2].reshape(2, 2)
heat_arr[:, 8:10] = pu_vecs[2].reshape(2, 2)
heat_arr[:, 10:] = ur.reshape(2, 2)
sns.heatmap(
heat_arr, ax=ax, annot=True, square=True, center=0.25,
cmap="Blues")
ax.invert_yaxis()
ax.set_ylim(2, 0)
ax.set_xlim(0, 12)
ax.vlines([k for k in range(2, 12, 2)], 2, 0, colors="r")
fig.savefig(
os.path.join(out_dir, str(key) + "_heatmap.png"))
out_dict["Summary Stats Rank"] = [
"Rot_Dist", "Rot_U_Dist", "Shuf_Dist", "Shuf_U_Dist", "Shuf_UR_Dist"]
out_dict["Summary Stats Rank"] += ["Rank" + b for b in base_list]
out_dict["Summary Stats Rank"] += ["Rot Rank" + b for b in base_list]
out_dict["Summary Stats Rank"] += ["Undo Rot Rank" + b for b in base_list]
out_dict["Summary Stats Rank"] += ["Shuf Rank" + b for b in base_list]
out_dict["Summary Stats Rank"] += ["Undo Shuf Rank" + b for b in base_list]
out_dict["Summary Stats Rank"] += ["Undo ShufR Rank" + b for b in base_list]
out_vec = to_rank(out_vec)
for key, vec in out_vec.items():
res, p_vecs, pu_vecs, ur = distance_between(
vec, key=key, measure=euc_dist)
out_dict["Rank Unit " + str(key)] = np.concatenate(
[res, p_vecs[0], p_vecs[1], pu_vecs[1], p_vecs[2], pu_vecs[2], ur])
print("Saving results to", os.path.join(
out_dir, str(tetrode) + "_obj.csv"))
save_mixed_dict_to_csv(out_dict, out_dir, str(tetrode) + "_obj.csv")
def main(args, config):
# Unpack out the cfg file into easier names
in_dir = config.get("Setup", "in_dir")
regex_filter = config.get("Setup", "regex_filter")
regex_filter = None if regex_filter == "None" else regex_filter
analysis_flags = json.loads(config.get("Setup", "analysis_flags"))
tetrode_list = json.loads(config.get("Setup", "tetrode_list"))
should_filter = config.getboolean("Setup", "should_filter")
seaborn_style = config.getboolean("Plot", "seaborn_style")
plot_order = json.loads(config.get("Plot", "plot_order"))
fixed_color = config.get("Plot", "path_color")
fixed_color = None if fixed_color == "None" else fixed_color
if len(fixed_color) > 1:
fixed_color = json.loads(fixed_color)
s_color = config.getboolean("Plot", "should_color")
plot_outname = config.get("Plot", "output_dirname")
dot_size = config.get("Plot", "dot_size")
dot_size = None if dot_size == "None" else int(dot_size)
summary_dpi = int(config.get("Plot", "summary_dpi"))
hd_predict = config.getboolean("Plot", "hd_predict")
output_format = config.get("Output", "output_format")
save_bin_data = config.getboolean("Output", "save_bin_data")
output_spaces = config.getboolean("Output", "output_spaces")
opt_end = config.get("Output", "optional_end")
max_units = int(config.get("Setup", "max_units"))
isi_bound = int(config.get("Params", "isi_bound"))
isi_bin_length = int(config.get("Params", "isi_bin_length"))
setup_logging(in_dir)
if output_format == "pdf":
matplotlib.use("pdf")
if seaborn_style:
sns.set(palette="colorblind")
else:
sns.set_style("ticks", {
'axes.spines.right': False,
'axes.spines.top': False
})
# Automatic extraction of files from starting dir onwards
container = NDataContainer(load_on_fly=True)
out_name = container.add_axona_files_from_dir(in_dir,
tetrode_list=tetrode_list,
recursive=True,
re_filter=regex_filter,
verbose=False,
unit_cutoff=(0, max_units))
container.setup()
if len(container) is 0:
print(
"Unable to find any files matching regex {}".format(regex_filter))
exit(-1)
# Show summary of place
if analysis_flags[0]:
place_cell_summary(container,
dpi=summary_dpi,
out_dirname=plot_outname,
filter_place_cells=should_filter,
filter_low_freq=should_filter,
opt_end=opt_end,
base_dir=in_dir,
output_format=output_format,
isi_bound=isi_bound,
isi_bin_length=isi_bin_length,
output=plot_order,
save_data=save_bin_data,
fixed_color=fixed_color,
point_size=dot_size,
color_isi=s_color,
burst_thresh=6,
hd_predict=hd_predict)
plt.close("all")
# Do numerical analysis of bursting
should_plot = analysis_flags[2]
if analysis_flags[1]:
import re
out_name = remove_extension(out_name) + "csv"
out_name = re.sub(r"file_list_", r"cell_stats_", out_name)
print("Computing cell stats to save to {}".format(out_name))
cell_classification_stats(in_dir,
container,
out_name,
should_plot=should_plot,
opt_end=opt_end,
output_spaces=output_spaces)
# Do PCA based analysis
if analysis_flags[3]:
print("Computing pca clustering")
pca_clustering(container, in_dir, opt_end=opt_end, s_color=s_color)