def read_data_frame_limited(fn, query_cols=[], max_rows=None):
''' Load a pandas DataFrame from an HDF5 file. If a column list is specified, only load the matching columns '''
with h5py.File(fn, 'r') as f:
column_names = f.attrs.get("column_names")
column_names = get_column_intersection(column_names, query_cols)
sz = f[column_names[0]].shape[0]
if max_rows:
sz = min(sz, max_rows)
df = p.DataFrame()
# Add the columns progressively to save memory
for name in column_names:
ds = f[name]
if has_levels(ds):
indices = ds[:sz]
uniques = get_levels(ds)
# This method of constructing of Categorical avoids copying the indices array
# which saves memory for big datasets
df[name] = p.Categorical(indices,
categories=uniques,
ordered=False,
fastpath=True)
else:
df[name] = p.Series(ds[:sz])
return df
def join(args, outs, chunk_defs, chunk_outs):
args_dict ={}
args_dict["bc_allow_indel"]=args.bc_allow_indel
args_dict["bc_max_error_allowed"]=args.bc_max_error_allowed
args_dict["bc_pseudo_count"]=args.bc_pseudo_count
args_dict["bc_use_mapping"]=args.bc_use_mapping
args_dict["bc_mapq"]=args.bc_mapq
args_dict["frag_no_merging"]=args.frag_no_merging
args_dict["frag_mapq"]=args.frag_mapq
args_dict["frag_pval"]=args.frag_pval
args_dict["frag_freq"]=args.frag_freq
fsummary = open(outs.summary, "w")
fsummary.write(safe_json.safe_jsonify(args_dict))
fsummary.close()
tk_bam.concatenate(out_file_name=outs.pos_sorted_bam, all_in_file_names=[chunk.pos_sorted_bam for chunk in chunk_outs])
tk_bam.index(outs.pos_sorted_bam)
outs.pos_sorted_bam_index = outs.pos_sorted_bam + '.bai'
bam_in = tk_bam.create_bam_infile(outs.pos_sorted_bam)
chroms = bam_in.references
barcode_whitelist = list(tk_seq.load_barcode_whitelist(args.barcode_whitelist))
barcode_whitelist.sort()
# Combine fragment csv files into a single h5 file
in_csv_files = [co.fragments+"_"+cd.tid+".csv" for (cd, co)
in zip(chunk_defs, chunk_outs) if os.path.exists(co.fragments+"_"+cd.tid+".csv")]
nfrags = 0
if len(in_csv_files) > 0:
bc_num_frags = defaultdict(int)
bc_num_reads = defaultdict(int)
bc_num_single_reads = defaultdict(int)
bc_num_lens = defaultdict(int)
temp_csv_barcodes = outs.barcodes+"_temp.csv"
nfrags = 0
for f in in_csv_files:
# TODO - sequentially append to fragments.h5 file to keep memory under control
# - handle multiple GEM groups properly.
# ensure the chroms column has string /categorical type in hdf5
# - same fixes for barcodes.h5 file
# handle 0-length outputs -- does that result in None file outs?
frag_in = p.read_csv(f, names=["tid", "start_pos", "end_pos", "bc_id", "num_reads"])
frag_in["obs_len"] = frag_in.end_pos - frag_in.start_pos
frag_in[frag_in.num_reads <= 1].obs_len = 1000
frag_in["est_len"] = np.maximum(1, frag_in["obs_len"] * (frag_in.num_reads + 1) / np.maximum(1, frag_in.num_reads - 1)).astype("int")
frag_in[frag_in.num_reads <= 1].est_len = 1000
barcode_seqs = []
molecule_ids = []
for (i, row) in frag_in.iterrows():
bc_num_frags[row.bc_id] += 1
bc_num_reads[row.bc_id] += row.num_reads
bc_num_lens[row.bc_id] += row.est_len
bc_wl_id = int(row.bc_id) % len(barcode_whitelist)
gg = int(row.bc_id) / len(barcode_whitelist) + 1
barcode_seq = "%s-%d" % (barcode_whitelist[bc_wl_id], gg)
barcode_seqs.append(barcode_seq)
molecule_ids.append(nfrags)
nfrags += 1
frag_in["bc"] = p.Categorical(barcode_seqs)
frag_in["chrom"] = p.Categorical.from_codes(frag_in.tid, chroms)
frag_in["molecule_id"] = molecule_ids
del frag_in["tid"]
del frag_in["bc_id"]
if len(frag_in) > 0:
tenkit.hdf5.append_data_frame(outs.fragments, frag_in)
with open(temp_csv_barcodes, "w") as csv_out:
csv_out.write("bc,bc_est_len,bc_linked_read_fraction,bc_linked_fragment_fraction,bc_mean_reads_per_fragment,bc_num_fragments,bc_num_reads\n")
for bc_id in range(len(barcode_whitelist)):
bc = barcode_whitelist[bc_id]+"-1"
if bc_id in bc_num_frags:
bc_est_len = bc_num_lens[bc_id]
bc_linked_read_fraction = 1.0 - bc_num_single_reads[bc_id]*1.0/bc_num_reads[bc_id]
bc_linked_fragment_fraction = 1.0 - bc_num_single_reads[bc_id]*1.0/bc_num_frags[bc_id]
bc_mean_reads_per_fragment = bc_num_reads[bc_id]*1.0/bc_num_frags[bc_id]
csv_out.write("%s,%d,%f,%f,%f,%d,%d\n" % (bc, bc_est_len, bc_linked_read_fraction, bc_linked_fragment_fraction, bc_mean_reads_per_fragment, bc_num_frags[bc_id], bc_num_reads[bc_id]))
if nfrags == 0:
outs.fragments = None
outs.barcodes = None
else:
tenkit.hdf5.create_tabix_index(outs.fragments, 'chrom', 'start_pos', 'end_pos')
df_barcodes = p.read_csv(temp_csv_barcodes)
tenkit.hdf5.append_data_frame(outs.barcodes, df_barcodes)
else:
outs.fragments = None
outs.barcodes= None
summary = {}
# Compute high-level BC summary metrics
# Load BC data
if outs.barcodes:
bc_df = tenkit.hdf5.read_data_frame(outs.barcodes)
fragment_df = tenkit.hdf5.read_data_frame(outs.fragments, query_cols=['bc', 'num_reads', 'est_len', 'chrom', 'start_pos'])
bc_df.sort('bc_num_reads', inplace=True)
# bin the bc counts and write a json histogram file
n_reads = bc_df.bc_num_reads.values
max_val = np.percentile(n_reads, 99.99) * 1.3
min_val = n_reads.min()
num_bins = 400
step = math.ceil((max_val - min_val)/num_bins)
bins = np.arange(min_val, max_val, step)
(hist, edges) = np.histogram(n_reads, bins=bins)
bc_count_hist = {int(edges[i]):hist[i] for i in range(len(bins)-1)}
# Summarize properties of n50 and n90 BC set
bc_df['cum_reads'] = np.cumsum(bc_df.bc_num_reads)
n50_read_thresh = sum(bc_df.bc_num_reads) * 0.5
n50_bcs = bc_df[bc_df.cum_reads > n50_read_thresh]
n50_fra = fragment_df[fragment_df.bc.isin(n50_bcs.bc)]
n50_stats = high_level_stats("n50", n50_fra, n50_bcs)
del n50_fra
n90_read_thresh = sum(bc_df.bc_num_reads) * 0.1
n90_bcs = bc_df[bc_df.cum_reads > n90_read_thresh]
n90_fra = fragment_df[fragment_df.bc.isin(n90_bcs.bc)]
n90_stats = high_level_stats("n90", n90_fra, n90_bcs)
del n90_fra
for (k,v) in n50_stats.iteritems():
summary[k] = v
for (k,v) in n90_stats.iteritems():
summary[k] = v
# Generate a fragment length histogram
fragment_df['len_bin'] = np.floor_divide(fragment_df.est_len.values, FRAG_LEN_HIST_BIN_SIZE).astype(int) * FRAG_LEN_HIST_BIN_SIZE
multi_read_frags = fragment_df[fragment_df.num_reads > 1]
len_bins = multi_read_frags.groupby(['len_bin']).apply(len)
del multi_read_frags
len_hist = {k:v for (k,v) in len_bins.iteritems()}
# Write fragment length hist to json
with open(outs.fragment_size, 'w') as fragment_size_file:
tenkit.safe_json.dump_numpy(len_hist, fragment_size_file)
# Estimate total DNA per partition by looking at hottest 1000 GEMs or GEMs w/ bc_mean_reads_per_fragment > 2, whichever is fewer
hot_bcs = bc_df[np.logical_and(bc_df.bc_mean_reads_per_fragment > 2.0, bc_df.bc_num_reads > 25)]
hot_bcs.sort('bc_mean_reads_per_fragment', inplace=True)
if len(hot_bcs) > 50:
hot_bcs = hot_bcs[-NUM_BCS_LOADING_ESTIMATE:]
summary['estimated_dna_per_partition'] = round(scipy.stats.tmean(hot_bcs.bc_est_len, scipy.percentile(hot_bcs.bc_est_len, (1,99))))
else:
summary['estimated_dna_per_partition'] = None
# Read-based effective diversity
reads = bc_df.bc_num_reads.values
sum_sq = (reads**2.0).sum()
effective_diversity = tk_stats.robust_divide((reads.sum()**2.0), float(sum_sq))
summary['effective_diversity_reads'] = effective_diversity
# Fragment-based effective diversity
fragments = bc_df.bc_num_fragments.values
sum_sq = (fragments**2.0).sum()
effective_diversity = tk_stats.robust_divide((fragments.sum()**2.0), float(sum_sq))
summary['effective_diversity_fragments'] = effective_diversity
else:
# No fragment_size file emitted
outs.fragment_size = None
n50_stats = high_level_stats("n50", None, None)
n90_stats = high_level_stats("n90", None, None)
for (k,v) in n50_stats.iteritems():
summary[k] = v
for (k,v) in n90_stats.iteritems():
summary[k] = v
bc_count_hist = {}
summary['estimated_dna_per_partition'] = None
summary['effective_diversity_reads'] = None
summary['effective_diversity_fragments'] = None
with open(outs.barcode_histogram, 'w') as barcode_hist_file:
tenkit.safe_json.dump_numpy(bc_count_hist, barcode_hist_file)
# Write summary to json
with open(outs.single_partition, 'w') as summary_file:
tenkit.safe_json.dump_numpy(summary, summary_file, pretty=True)