File size: 20,604 Bytes
fe41391
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
#!/usr/bin/env python
#####################################################################################
# rust_codon, Produces RUST metagene profile of codons
# Copyright (C) 2015 Patrick O'Connor

# This program is free software: you can redistribute it and/or modify
# it under the terms of the GNU General Public License as published by
# the Free Software Foundation, either version 3 of the License, or
# (at your option) any later version.

# This program is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
# GNU General Public License for more details.

# You should have received a copy of the GNU General Public License
# along with this program.  If not, see <https://www.gnu.org/licenses/>.
#####################################################################################


import os, re, pysam, sys, math, argparse
from RUST.methods import *

try:
    import matplotlib as mpl

    mpl.use("Agg")
    import matplotlib.pyplot as plt
    from pylab import MaxNLocator
except:
    pass


def RUST_metagene_plot(infileopen36, ax36):
    infileopen36.seek(0)
    infileopen36.readline()
    while 1:
        line = infileopen36.readline()
        linesplit = line.split(",")
        if len(linesplit) == 1:
            break
        codon = linesplit[0]
        coverage = list(map(float, linesplit[1:]))
        coverage_a = coverage[0]
        if coverage_a == 0:
            continue
        coverage_n = [n / coverage_a for n in coverage[1:]]
        log2_values = [math.log(n, 2) if n != 0 else float("-inf") for n in coverage_n]
        ax36.plot(log2_values, color="gray")

    line = infileopen36.readline()
    linesplit = line.split(",")
    if "NA" not in line:
        coverage = map(float, linesplit[2:])
        ax2 = ax36.twinx()
        ax2.plot(coverage, color="blue")
        for tl in ax2.get_yticklabels():
            tl.set_color("blue")
            tl.set_rotation(0)

        ax2.yaxis.set_major_locator(MaxNLocator(3))
        ax2.set_ylim(0, 1.0)
        ax2.set_ylim(-2, 1.0)
        ax2.set_yticks([0, 1], minor=False)
        ax2.set_yticklabels(["0", "1"])
        ax2.set_ylabel("Kullback-Leibler divergence", color="blue")

    ax36.set_xticks([5, 10, 15, 20, 25, 30, 35, 40, 45, 50, 55])
    ax36.set_xticklabels([-35, -30, -25, -20, -15, -10, -5, 0, 5, 10, 15])
    ax36.set_xlabel("distance from A-site [codon]")
    ax36.set_ylabel("Codon RUST ratio (observed/expected), log2")
    ax36.axvline(40, color="red")


def A_site_plot(infileopen35, dict_codon75, axis_Asite53, loc2):
    codon_to_amino_dict = {}
    amino_to_codons_dict = {}
    for amino_acid, codons in dict_codon75.items():
        for codon in codons:
            codon_to_amino_dict[codon] = amino_acid
            amino_to_codons_dict.setdefault(amino_acid, []).append(codon)

    list1 = []
    list2 = []
    infileopen35.seek(0)
    infileopen35.readline()
    dict_amino_value = {}
    for line in infileopen35:
        linesplit = line.split(",")
        if len(linesplit) == 1:
            break
        codon = linesplit[0]
        if codon in ["TAA", "TAG", "TGA"]:
            continue
        list1.append(linesplit[0])
        coverage = list(map(float, linesplit[1:]))
        coverage_a = coverage[0]
        coverage_n = [n / coverage_a for n in coverage[1:]]
        list2.append(float(coverage_n[loc2]))

        amino = codon_to_amino_dict[linesplit[0]]
        if amino in dict_amino_value:
            dict_amino_value[codon_to_amino_dict[linesplit[0]]].append(
                float(coverage_n[loc2])
            )
        else:
            dict_amino_value[codon_to_amino_dict[linesplit[0]]] = [
                float(coverage_n[loc2])
            ]

    list_amino_sorted = []
    for key, value in dict_amino_value.items():
        list_amino_sorted.append((mean_value(value), key))
    list_amino_sorted.sort()

    A_site_value_norm = [n / min(list2) for n in list2]
    list3 = list(zip(A_site_value_norm, list1))
    list3.sort()
    A_site_value_norm_dict = {}
    for tupel in list3:
        A_site_value_norm_dict[tupel[1]] = tupel[0]

    used_codons = []
    xloc = []
    xtick_label = []
    n1 = 0

    for _, amino_acid in list_amino_sorted:
        if amino_acid in used_codons:
            continue
        used_codons.append(amino_acid)
        n1 += 1  # len(dict_list_codon[amino_acid])

        xloc.append(n1)
        for amino_acid_codon in amino_to_codons_dict[amino_acid]:
            axis_Asite53.scatter(
                n1,
                A_site_value_norm_dict[amino_acid_codon],
                color="gray",
                s=50,
                edgecolor="gray",
            )
        xtick_label.append(amino_acid)

    axis_Asite53.set_xticks(xloc)
    axis_Asite53.set_xticklabels(xtick_label, rotation=90)
    for tick in axis_Asite53.get_xticklabels():
        if tick.get_text() in ["Phe", "Tyr", "Trp"]:
            a2 = tick.set_backgroundcolor("lightgreen")  # (dict(facecolor = "red"))
            # tick.set_color("white")
        if tick.get_text() in ["Val", "Ala", "Leu", "Met", "Ile"]:
            tick.set_backgroundcolor("lightgrey")
            # tick.set_color("white")
        if tick.get_text() in ["Ser", "Asn", "Thr", "Gln"]:
            tick.set_backgroundcolor("ForestGreen")
            tick.set_color("white")

        if tick.get_text() in ["His", "Lys", "Arg"]:
            tick.set_backgroundcolor("blue")
            tick.set_color("white")
        if tick.get_text() in ["Glu", "Asp"]:
            tick.set_backgroundcolor("red")
            tick.set_color("white")
    axis_Asite53.set_xlim(0, n1 + 1)
    axis_Asite53.set_ylabel("A-site codon RUST ratio")

    red = mpl.patches.Rectangle((0, 0), 1, 1, fc="r")
    blue = mpl.patches.Rectangle((0, 0), 1, 1, fc="b")
    fgreen = mpl.patches.Rectangle((0, 0), 1, 1, fc="ForestGreen")
    lgreen = mpl.patches.Rectangle((0, 0), 1, 1, fc="lightGreen")
    grey = mpl.patches.Rectangle((0, 0), 1, 1, fc="lightgrey")

    axis_Asite53.legend(
        [red, grey, lgreen, blue, fgreen],
        ["acidic", "aliphatic", "aromatic", "basic", "polar\nuncharged"],
        bbox_to_anchor=(0, 0, 0.8, 1.12),
        ncol=3,
    )


def main(args):

    universal_code = {
        "Ala": ["GCT", "GCC", "GCG", "GCA"],
        "Gly": ["GGT", "GGC", "GGG", "GGA"],
        "Pro": ["CCT", "CCC", "CCG", "CCA"],
        "Thr": ["ACT", "ACC", "ACG", "ACA"],
        "Val": ["GTT", "GTC", "GTG", "GTA"],
        "Ser": ["TCT", "TCC", "TCG", "TCA", "AGT", "AGC"],
        "Arg": ["CGT", "CGC", "CGG", "CGA", "AGG", "AGA"],
        "Leu": ["CTT", "CTC", "CTG", "CTA", "TTG", "TTA"],
        "Phe": ["TTT", "TTC"],
        "Asn": ["AAT", "AAC"],
        "Lys": ["AAG", "AAA"],
        "Asp": ["GAT", "GAC"],
        "Glu": ["GAG", "GAA"],
        "His": ["CAT", "CAC"],
        "Gln": ["CAG", "CAA"],
        "Ile": ["ATT", "ATC", "ATA"],
        "Met": ["ATG"],
        "Tyr": ["TAT", "TAC"],
        "Cys": ["TGT", "TGC"],
        "Trp": ["TGG"],
        "Stop": ["TGA", "TAG", "TAA"],
    }

    mRNA_sequences = args.transcriptome  # path to fastq file of transcripts
    in_seq_handle = open(mRNA_sequences)
    cds_start_dict = {}
    cds_end_dict = {}
    seq_dict = {}
    for line in in_seq_handle:
        if line[0] != ">":
            seq_dict.setdefault(transcript, "")
            seq_dict[transcript] += line[:-1]
            continue
        try:
            transcript_split = line[:-1].split("\t")
            transcript = transcript_split[0][1:]
            cds_start_dict[transcript] = int(transcript_split[1])
            cds_end_dict[transcript] = int(transcript_split[2])
        except:
            pass
    in_seq_handle.close()

    offset = args.offset
    readlen_range = args.lengths

    readlen_rangesplit = readlen_range.split(":")
    if len(readlen_rangesplit) == 1:
        accepted_read_lengths = [int(readlen_rangesplit[0])]
        length_values = "%s" % int(readlen_rangesplit[0])
    elif len(readlen_rangesplit) == 2:
        accepted_read_lengths = [
            readlen
            for readlen in range(
                int(readlen_rangesplit[0]), int(readlen_rangesplit[1]) + 1
            )
        ]
        length_values = "%s_%s" % (
            int(readlen_rangesplit[0]),
            int(readlen_rangesplit[1]),
        )
    else:
        stop_err(
            "Lengths of footprints parameter not in correct format, it should be either colon seperated with the second value greater or equal to the first, (28:32) or a single interger (31)"
        )
    if len(accepted_read_lengths) == 0:
        stop_err(
            "Lengths of footprints parameter not in correct format, it should be either colon seperated with the second value greater or equal to the first, (28:32) or a single interger (31)"
        )

    nts = ["A", "G", "C", "T"]
    aligments_A1 = pysam.Samfile(
        args.alignment, "rb"
    )  # path to aligments in bam format

    codon_enrichment_dict = {}
    codon_enrichment_expected_dict = {}
    for nt in nts:
        for nt2 in nts:
            for nt3 in nts:
                codon = "%s%s%s" % (nt, nt2, nt3)
                codon_enrichment_dict[codon] = {}
                codon_enrichment_expected_dict[codon] = []
                for number in range(0, 60, 1):
                    codon_enrichment_dict[codon][number] = [0.0, 0.0]

    list_transcripts = seq_dict.keys()
    number_transcripts = 0
    list_10_percentile = []
    for value in range(1, 10):
        list_10_percentile.append((len(list_transcripts) * value) / 10)
    for transcript in list_transcripts:
        number_transcripts += 1
        if number_transcripts in list_10_percentile:
            sys.stdout.write(
                "%s percent\n"
                % ((list_10_percentile.index(number_transcripts) + 1) * 10)
            )

        try:  # use supplied CDS annotation
            cds_start = cds_start_dict[transcript]
            cds_end = cds_end_dict[transcript]
            if cds_end < cds_start:
                raise Exception
        except Exception:  # find longest ORF
            transcript_seq = seq_dict[transcript]
            cds_start = -1
            start_post = []
            end_post = []
            for match in re.finditer(r"(?=(%s))" % re.escape("ATG"), transcript_seq):
                start_post.append(match.start())
            for match in re.finditer(r"(?=(%s))" % re.escape("TAG"), transcript_seq):
                end_post.append(match.start())
            for match in re.finditer(r"(?=(%s))" % re.escape("TAA"), transcript_seq):
                end_post.append(match.start())
            for match in re.finditer(r"(?=(%s))" % re.escape("TGA"), transcript_seq):
                end_post.append(match.start())

            end_post.sort()
            len_max_orf = 0
            for value in start_post:
                for value2 in end_post:
                    if value < value2:
                        if value % 3 == value2 % 3:
                            len_orf = value2 - value
                            if len_orf > len_max_orf:
                                cds_start = value
                                cds_end = value2 + 3
                                len_max_orf = len_orf
                            break
            if cds_start == -1:
                # sys.stdout.write( '%s, AUG codon not found\n'%transcript  )
                continue

        elongation_region_all = seq_dict[transcript][cds_start:cds_end]
        elongation_region_part = elongation_region_all[
            120:-60
        ]  # first 120 and last 60 nt are not used
        # peptide_sequence        = elongation_region_all.translate()

        if len(elongation_region_part) % 3 != 0:
            # sys.stdout.write( '%s, CDS not divisible by 3\n'%transcript  )
            continue

        profile_list = [
            0.0 for n in range(cds_start + 120, cds_end - 60)
        ]  # records ribo-seq profile
        if len(profile_list) < 50:
            # sys.stdout.write( '%s, ORF too short\n'%transcript  )
            continue
        all_reads = aligments_A1.fetch(transcript)

        len_elongation_region = len(profile_list)
        for read in all_reads:
            readlen = read.qlen
            if readlen not in accepted_read_lengths:
                continue  # selection of read of acceptable length
            A_site = read.pos + offset - cds_start - 120  # addition of offset
            if len_elongation_region > A_site > -1:
                profile_list[A_site] += 1

        average_gene_density = float(sum(profile_list)) / len(
            profile_list
        )  # average gene density calculated
        if average_gene_density != 0:
            num_codon = len(
                [
                    1
                    for number88 in range(0, len(profile_list), 3)
                    if (
                        (
                            profile_list[number88]
                            + profile_list[number88 + 1]
                            + profile_list[number88 + 2]
                        )
                        / 3
                    )
                    > average_gene_density
                ]
            )
            # number of codons that exceed average gene density
            expected_codon_density = float(num_codon) / (
                len(profile_list) / 3
            )  # expected enrichment value

            codon_start = 0
            for sliding_w_n in range(
                0, len(elongation_region_part), 3
            ):  # sliding window using increments of 3 nts
                codon_window = str(
                    elongation_region_all[codon_start : codon_start + 180]
                )  # 60 codon window,
                if len(set(codon_window) - set(["A", "T", "G", "C"])) != 0:
                    codon_start += 3
                    continue

                if (
                    profile_list[sliding_w_n]
                    + profile_list[sliding_w_n + 1]
                    + profile_list[sliding_w_n + 2]
                ) / 3 > average_gene_density:
                    for number in range(0, 60):
                        codon = codon_window[number * 3 : (number + 1) * 3]
                        codon_enrichment_dict[codon][number][0] += 1
                        codon_enrichment_dict[codon][number][1] += 1
                else:
                    for number in range(0, 60):
                        codon = codon_window[number * 3 : (number + 1) * 3]
                        codon_enrichment_dict[codon][number][0] += 1

                codon = codon_window[120:123]  # corresponds to A-site codon
                codon_enrichment_expected_dict[codon].append(expected_codon_density)
                codon_start += 3

    if not os.path.exists(args.Path):
        os.mkdir(args.Path)
    alignment_filename = args.alignment.split("/")[-1]
    outfile = open(
        "%s/RUST_codon_file_%s_%s_%s"
        % (args.Path, alignment_filename, args.offset, length_values),
        "w",
    )
    outfile.write("codon, expected value")
    for number106 in range(-40, 20):
        outfile.write(", %s" % number106)
    outfile.write("\n")

    list_codons = []
    codons = list(codon_enrichment_dict)
    codons.sort()
    rust_expected = []
    rust_observed_metafootprint = []
    for codon in codons:
        if codon in list_codons:
            continue
        if codon in ["TAA", "TGA", "TAG"]:
            continue
        list_codons.append(codon)
        outfile.write("%s" % codon)
        if codon_enrichment_expected_dict[codon] != []:
            outfile.write(", %s" % mean_value(codon_enrichment_expected_dict[codon]))
        list_data = []
        for number in range(0, 60):
            if codon_enrichment_dict[codon][number][0] != 0:
                outfile.write(
                    ", %s"
                    % (
                        codon_enrichment_dict[codon][number][1]
                        / codon_enrichment_dict[codon][number][0]
                    )
                )
                list_data.append(
                    codon_enrichment_dict[codon][number][1]
                    / codon_enrichment_dict[codon][number][0]
                )
            else:
                outfile.write(", 0")
                list_data.append(0)
        outfile.write("\n")
        rust_expected.append(mean_value(codon_enrichment_expected_dict[codon]))
        rust_observed_metafootprint.append(list_data)

    rust_expected_sum = sum(rust_expected)
    q_values = [n / rust_expected_sum for n in rust_expected]

    shannon_values = []
    for loc_i in range(60):
        rust_observed = [n[loc_i] for n in rust_observed_metafootprint]
        rust_observed_sum = sum(rust_observed)
        rust_observed_min = min(rust_observed)
        if rust_observed_min == 0:
            shannon_values.append("NA")
        else:
            p_values = [n / rust_observed_sum for n in rust_observed]
            shannon = []
            list_normalised = []  ####
            for p_value, q_value in zip(p_values, q_values):
                shannon.append(abs(p_value * math.log((p_value / q_value), 2)))
                list_normalised.append(p_value / q_value)  ####
            shannon_values.append(sum(shannon))

    outfile.write("\nKullback Leibler divergence,")
    for value in shannon_values:
        outfile.write(", %s" % value)
    outfile.close()

    try:
        mpl.rcParams["xtick.direction"] = "out"
        mpl.rcParams["ytick.direction"] = "out"
        mpl.rcParams["legend.fontsize"] = 10
        mpl.rcParams["ytick.labelsize"] = 10
        mpl.rcParams["xtick.labelsize"] = 10
        mpl.rcParams["font.size"] = 10
        mpl.rcParams["axes.titlesize"] = 10
        mpl.rcParams["legend.frameon"] = 0
        mpl.rcParams["axes.axisbelow"] = False
        mpl.rcParams["xtick.major.pad"] = 2.0
        mpl.rcParams["ytick.major.pad"] = 2
        mpl.rcParams["xtick.major.size"] = 2.0
        mpl.rcParams["ytick.major.size"] = 2
        mpl.rcParams["axes.linewidth"] = 0.5
        mpl.rcParams["ytick.major.width"] = 0.25
        mpl.rcParams["xtick.major.width"] = 0.25
        mpl.rcParams["lines.linewidth"] = 1
        mpl.rcParams["legend.borderpad"] = 0.01
        mpl.rcParams["legend.labelspacing"] = 0.05
        mpl.rcParams["legend.columnspacing"] = 0.5
        mpl.rcParams["legend.borderaxespad"] = 0.15
        mpl.rcParams["legend.handlelength"] = 1

        fig = plt.figure(figsize=(6.69, 6.0))
        infileopen = open(
            "%s/RUST_codon_file_%s_%s_%s"
            % (args.Path, alignment_filename, args.offset, length_values)
        )
        ax1_metafootprint = fig.add_subplot(111)
        RUST_metagene_plot(infileopen, ax1_metafootprint)
        plt.savefig(
            "%s/RUST_codon_metafootprint_%s_%s_%s.png"
            % (args.Path, alignment_filename, args.offset, length_values)
        )
        plt.clf()

        infileopen = open(
            "%s/RUST_codon_file_%s_%s_%s"
            % (args.Path, alignment_filename, args.offset, length_values)
        )
        ax1codon_Asite = fig.add_subplot(111)
        A_site_plot(infileopen, universal_code, ax1codon_Asite, 40)
        plt.savefig(
            "%s/A_site_%s_%s_%s.png"
            % (args.Path, alignment_filename, args.offset, length_values)
        )

    except:
        sys.stdout.write("Error producing images\n")


if __name__ == "__main__":
    parser = argparse.ArgumentParser(
        description="Produces RUST metagene profile of codons"
    )
    parser.add_argument("--version", action="version", version="%(prog)s 1.2")

    parser.add_argument(
        "-t",
        "--transcriptome",
        help="fasta file of transcripts, CDS start and end may be provided on description line using tab separation e.g. >NM_0001  10  5000, otherwise it searches for longest ORF"
        ", required=True",
    )
    parser.add_argument(
        "-a",
        "--alignment",
        help="sorted bam file of transcriptome alignments",
        required=True,
    )
    parser.add_argument("-o", "--offset", help="nucleotide offset to A-site", type=int)
    parser.add_argument(
        "-l",
        "--lengths",
        help="lengths of footprints included, for example 28:32 is 28,29,30,31,32",
    )
    parser.add_argument(
        "-P", "--Path", help='path to outputfile, default is "codon"', default="codon"
    )
    args = parser.parse_args(None)

    main(args)