varvamp 1.1.3__py3-none-any.whl → 1.2.1__py3-none-any.whl

varvamp/__init__.py

@@ -1,3 +1,3 @@
 """Tool to design amplicons for highly variable virusgenomes"""
 _program = "varvamp"
-__version__ = "1.1.3"
+__version__ = "1.2.1"

varvamp/command.py

@@ -89,6 +89,13 @@ def get_args(sysargs):
             type=int,
             default=1
         )
+        par.add_argument(
+            "--name",
+            help="name of the scheme",
+            metavar="varVAMP",
+            type=str,
+            default="varVAMP"
+        )
     for par in (SINGLE_parser, TILED_parser):
         par.add_argument(
             "-ol",
@@ -261,10 +268,10 @@ def shared_workflow(args, log_file):
         ambiguous_consensus,
         alignment_cleaned
     )
-    for type, primer_candidates in [("+", left_primer_candidates), ("-", right_primer_candidates)]:
+    for primer_type, primer_candidates in [("+", left_primer_candidates), ("-", right_primer_candidates)]:
         if not primer_candidates:
             logging.raise_error(
-                f"no {type} primers found.\n",
+                f"no {primer_type} primers found.\n",
                 log_file,
                 exit=True
             )
@@ -314,9 +321,9 @@ def single_and_tiled_shared_workflow(args, left_primer_candidates, right_primer_
 
     if args.database is not None:
         # create blast query
-        query_path = blast.create_BLAST_query(all_primers, amplicons, data_dir)
+        query_path = blast.create_BLAST_query(amplicons, data_dir)
         # perform primer blast
-        amplicons, off_target_amplicons = blast.primer_blast(
+        amplicons = blast.primer_blast(
             data_dir,
             args.database,
             query_path,
@@ -326,23 +333,21 @@ def single_and_tiled_shared_workflow(args, left_primer_candidates, right_primer_
             log_file,
             mode="single_tiled"
         )
-    else:
-        off_target_amplicons = []
 
-    return all_primers, amplicons, off_target_amplicons
+    return all_primers, amplicons
 
 
-def single_workflow(args, amplicons, all_primers, log_file):
+def single_workflow(args, amplicons, log_file):
     """
     workflow part specific for single mode
     """
 
-    amplicon_scheme = scheme.find_single_amplicons(amplicons, all_primers, args.report_n)
+    amplicon_scheme = scheme.find_single_amplicons(amplicons, args.report_n)
     logging.varvamp_progress(
         log_file,
         progress=0.9,
         job="Finding amplicons with low penalties.",
-        progress_text=f"{len(amplicon_scheme[0])} amplicons."
+        progress_text=f"{len(amplicon_scheme)} amplicons."
     )
 
     return amplicon_scheme
@@ -359,8 +364,7 @@ def tiled_workflow(args, amplicons, left_primer_candidates, right_primer_candida
     # search for amplicon scheme
     coverage, amplicon_scheme = scheme.find_best_covering_scheme(
         amplicons,
-        amplicon_graph,
-        all_primers
+        amplicon_graph
     )
 
     # check for dimers
@@ -377,12 +381,13 @@ def tiled_workflow(args, amplicons, left_primer_candidates, right_primer_candida
         reporting.write_dimers(results_dir, dimers_not_solved)
 
     # evaluate coverage
+    # ATTENTION: Genome coverage of the scheme might still change slightly through resolution of primer dimers, but this potential, minor inaccuracy is currently accepted.
     percent_coverage = round(coverage/len(ambiguous_consensus)*100, 2)
     logging.varvamp_progress(
         log_file,
         progress=0.9,
         job="Creating amplicon scheme.",
-        progress_text=f"{percent_coverage} % total coverage with {len(amplicon_scheme[0]) + len(amplicon_scheme[1])} amplicons"
+        progress_text=f"{percent_coverage} % total coverage with {len(amplicon_scheme)} amplicons"
     )
     if percent_coverage < 70:
         logging.raise_error(
@@ -450,9 +455,9 @@ def qpcr_workflow(args, data_dir, alignment_cleaned, ambiguous_consensus, majori
     # run blast if db is given
     if args.database is not None:
         # create blast query
-        query_path = blast.create_BLAST_query_qpcr(qpcr_scheme_candidates, data_dir)
+        query_path = blast.create_BLAST_query(qpcr_scheme_candidates, data_dir, mode="qpcr")
         # perform primer blast
-        amplicons, off_target_amplicons = blast.primer_blast(
+        qpcr_scheme_candidates = blast.primer_blast(
             data_dir,
             args.database,
             query_path,
@@ -470,9 +475,6 @@ def qpcr_workflow(args, data_dir, alignment_cleaned, ambiguous_consensus, majori
             log_file,
             exit=True
         )
-    # report potential blast warnings
-    if args.database is not None:
-        blast.write_BLAST_warning(off_target_amplicons, final_schemes, log_file)
     logging.varvamp_progress(
         log_file,
         progress=0.9,
@@ -482,13 +484,13 @@ def qpcr_workflow(args, data_dir, alignment_cleaned, ambiguous_consensus, majori
     return probe_regions, final_schemes
 
 
-def main(sysargs=sys.argv[1:]):
+def main():
     """
     main varvamp workflow
     """
 
     # start varVAMP
-    args = get_args(sysargs)
+    args = get_args(sys.argv[1:])
     if not args.verbose:
         sys.stdout = open(os.devnull, 'w')
     start_time = datetime.datetime.now()
@@ -501,14 +503,26 @@ def main(sysargs=sys.argv[1:]):
     alignment_cleaned, majority_consensus, ambiguous_consensus, primer_regions, left_primer_candidates, right_primer_candidates = shared_workflow(args, log_file)
 
     # write files that are shared in all modes
-    reporting.write_regions_to_bed(primer_regions, data_dir)
+    reporting.write_regions_to_bed(primer_regions, args.name, data_dir)
     reporting.write_alignment(data_dir, alignment_cleaned)
-    reporting.write_fasta(data_dir, "majority_consensus", majority_consensus)
-    reporting.write_fasta(results_dir, "ambiguous_consensus", ambiguous_consensus)
+    reporting.write_fasta(data_dir, f"majority_consensus", f"{args.name}_consensus",majority_consensus)
+    reporting.write_fasta(results_dir, f"ambiguous_consensus", f"{args.name}_consensus", ambiguous_consensus)
+
+    # Functions called from here on return lists of amplicons that are refined step-wise into final schemes.
+    # These lists that are passed between functions and later used for reporting consist of dictionary elemnts,
+    # which represent individual amplicons. A minimal amplicon dict could take the form:
+    # {
+    #     "id": amplicon_name,
+    #     "penalty": amplicon_cost,
+    #     "length": amplicon_length,
+    #     "LEFT": [left primer data],
+    #     "RIGHT": [right primer data]
+    # }
+    # to which different functions may add additional information.
 
     # SINGLE/TILED mode
     if args.mode == "tiled" or args.mode == "single":
-        all_primers, amplicons, off_target_amplicons = single_and_tiled_shared_workflow(
+        all_primers, amplicons = single_and_tiled_shared_workflow(
             args,
             left_primer_candidates,
             right_primer_candidates,
@@ -519,7 +533,6 @@ def main(sysargs=sys.argv[1:]):
             amplicon_scheme = single_workflow(
                 args,
                 amplicons,
-                all_primers,
                 log_file
             )
         elif args.mode == "tiled":
@@ -533,24 +546,33 @@ def main(sysargs=sys.argv[1:]):
                 log_file,
                 results_dir
             )
-        if args.database is not None:
-            blast.write_BLAST_warning(off_target_amplicons, amplicon_scheme, log_file)
+
         # write files
-        reporting.write_all_primers(data_dir, all_primers)
+
+        if args.mode == "tiled":
+            # assign amplicon numbers from 5' to 3' along the genome
+            amplicon_scheme.sort(key=lambda x: x["LEFT"][1])
+        else:
+            # make sure amplicons with no off-target products and with low penalties get the lowest numbers
+            amplicon_scheme.sort(key=lambda x: (x.get("off_targets", False), x["penalty"]))
+        reporting.write_all_primers(data_dir, args.name, all_primers)
         reporting.write_scheme_to_files(
             results_dir,
             amplicon_scheme,
             ambiguous_consensus,
-            args.mode
+            args.name,
+            args.mode,
+            log_file
         )
         reporting.varvamp_plot(
             results_dir,
             alignment_cleaned,
             primer_regions,
+            args.name,
             all_primers=all_primers,
             amplicon_scheme=amplicon_scheme,
         )
-        reporting.per_base_mismatch_plot(results_dir, amplicon_scheme, args.threshold)
+        reporting.per_base_mismatch_plot(results_dir, amplicon_scheme, args.threshold, args.name)
 
     # QPCR mode
     if args.mode == "qpcr":
@@ -564,17 +586,22 @@ def main(sysargs=sys.argv[1:]):
             right_primer_candidates,
             log_file
         )
+
         # write files
-        reporting.write_regions_to_bed(probe_regions, data_dir, "probe")
-        reporting.write_qpcr_to_files(results_dir, final_schemes, ambiguous_consensus)
+
+        # make sure amplicons with no off-target products and with low penalties get the lowest numbers
+        final_schemes.sort(key=lambda x: (x.get("off_targets", False), x["penalty"]))
+        reporting.write_regions_to_bed(probe_regions, args.name, data_dir, "probe")
+        reporting.write_qpcr_to_files(results_dir, final_schemes, ambiguous_consensus, args.name, log_file)
         reporting.varvamp_plot(
             results_dir,
             alignment_cleaned,
             primer_regions,
+            args.name,
             probe_regions=probe_regions,
             amplicon_scheme=final_schemes
         )
-        reporting.per_base_mismatch_plot(results_dir, final_schemes, args.threshold, mode="QPCR")
+        reporting.per_base_mismatch_plot(results_dir, final_schemes, args.threshold, args.name, mode="QPCR")
 
     # varVAMP finished
     logging.varvamp_progress(log_file, progress=1, start_time=start_time)

varvamp/scripts/blast.py

@@ -29,41 +29,24 @@ def check_BLAST_installation(log_file):
         logging.raise_error("BLASTN is not installed", log_file, exit=True)
 
 
-def create_BLAST_query(all_primers, amplicons, data_dir):
+def create_BLAST_query(amplicons, data_dir, mode="single_tiled"):
     """
-    create a query for the BLAST search (tiled, single mode)
+    create a query for the BLAST search
     """
-    already_written = []
-
     query_path = os.path.join(data_dir, "BLAST_query.fasta")
-    with open(query_path, "w") as query:
-        for amp in amplicons:
-            fw_primer, rv_primer = amplicons[amp][2], amplicons[amp][3]
-            if fw_primer not in already_written:
-                print(f">{fw_primer}\n{all_primers['+'][fw_primer][0]}", file=query)
-                already_written.append(fw_primer)
-            if rv_primer not in already_written:
-                print(f">{rv_primer}\n{all_primers['-'][rv_primer][0]}", file=query)
-                already_written.append(rv_primer)
-
-    return query_path
-
-
-def create_BLAST_query_qpcr(qpcr_scheme_candidates, data_dir):
-    """
-    create a query for the BLAST search (qpcr mode)
-    """
-    already_written = []
+    if mode == "single_tiled":
+        primer_types = ["LEFT", "RIGHT"]
+    elif mode == "qpcr":
+        primer_types = ["PROBE", "LEFT", "RIGHT"]
+    already_written = set()
 
-    query_path = os.path.join(data_dir, "BLAST_query.fasta")
     with open(query_path, "w") as query:
-        for amp in qpcr_scheme_candidates:
-            for primer_type in ["PROBE", "LEFT", "RIGHT"]:
-                name = f"{primer_type}_{qpcr_scheme_candidates[amp][primer_type][1]}_{qpcr_scheme_candidates[amp][primer_type][2]}"
-                if name in already_written:
-                    continue
-                print(f">{name}\n{qpcr_scheme_candidates[amp][primer_type][0]}", file=query)
-                already_written.append(name)
+        for amp in amplicons:
+            for primer_type in primer_types:
+                name = f"{primer_type}_{amp[primer_type][1]}_{amp[primer_type][2]}"
+                if name not in already_written:
+                    print(f">{name}\n{amp[primer_type][0]}", file=query)
+                    already_written.add(name)
     return query_path
 
 
@@ -168,21 +151,24 @@ def predict_non_specific_amplicons_worker(amp, blast_df, max_length, mode):
     """
     Worker function to predict unspecific targets for a single amplicon.
     """
-    name, data = amp
     # get correct primers
     if mode == "single_tiled":
-        primers = [data[2], data[3]]
+        primer_types = ["LEFT", "RIGHT"]
     elif mode == "qpcr":
-        primers = []
-        for primer_type in ["PROBE", "LEFT", "RIGHT"]:
-            primers.append(f"{primer_type}_{data[primer_type][1]}_{data[primer_type][2]}")
+        primer_types = ["PROBE", "LEFT", "RIGHT"]
+    primers = []
+    for primer_type in primer_types:
+        primers.append(f"{primer_type}_{amp[primer_type][1]}_{amp[primer_type][2]}")
     # subset df for primers
     df_amp_primers = blast_df[blast_df["query"].isin(primers)]
     # sort by reference and ref start
     df_amp_primers_sorted = df_amp_primers.sort_values(["ref", "ref_start"])
     # check for off-targets for specific primers
     if check_off_targets(df_amp_primers_sorted, max_length, primers):
-        return name
+        amp["off_targets"] = True
+    else:
+        amp["off_targets"] = False
+    return amp
 
 
 def predict_non_specific_amplicons(amplicons, blast_df, max_length, mode, n_threads):
@@ -190,22 +176,16 @@ def predict_non_specific_amplicons(amplicons, blast_df, max_length, mode, n_thre
     Main function to predict unspecific targets within a size range and give
     these primers a high penalty. Uses multiprocessing for parallelization.
     """
-    off_targets = []
     # process amplicons concurrently
     with multiprocessing.Pool(processes=n_threads) as pool:
-        amp_items = amplicons.items()
-        results = pool.starmap(predict_non_specific_amplicons_worker, [(amp, blast_df, max_length, mode) for amp in amp_items])
-    # check results
-    for off_target in results:
-        if off_target is None:
-            continue
-        off_targets.append(off_target)
-        if mode == "single_tiled":
-            amplicons[off_target][5] = amplicons[off_target][5] + config.BLAST_PENALTY
-        elif mode == "qpcr":
-            amplicons[off_target]["penalty"] = amplicons[off_target]["penalty"] + config.BLAST_PENALTY
-
-    return off_targets, amplicons
+        annotated_amps = [
+            result for result in pool.starmap(
+                predict_non_specific_amplicons_worker,
+                [(amp, blast_df, max_length, mode) for amp in amplicons]
+            ) if result is not None
+        ]
+    n_off_targets = sum(amp["off_targets"] for amp in annotated_amps)
+    return n_off_targets, annotated_amps
 
 
 def primer_blast(data_dir, db, query_path, amplicons, max_length, n_threads, log_file, mode):
@@ -237,14 +217,17 @@ def primer_blast(data_dir, db, query_path, amplicons, max_length, n_threads, log
 
     blast_df = parse_and_filter_BLAST_output(blast_out)
     print("Predicting non-specific amplicons...")
-    off_target_amplicons, amplicons = predict_non_specific_amplicons(
+    n_off_targets, amplicons = predict_non_specific_amplicons(
         amplicons,
         blast_df,
         max_length,
         mode,
         n_threads
     )
-    success_text = f"varVAMP successfully predicted non-specific amplicons:\n\t> {len(off_target_amplicons)}/{len(amplicons)} amplicons could produce amplicons with the blast db.\n\t> raised their amplicon penalty by {config.BLAST_PENALTY}"
+    if n_off_targets > 0:
+        success_text = f"varVAMP predicted non-specific amplicons:\n\t> {n_off_targets}/{len(amplicons)} amplicons could produce amplicons with the blast db.\n\t> will attempt to avoid them in the final list of amplicons"
+    else:
+        success_text = f"NO off-target amplicons found with the blast db and a total of {len(amplicons)} amplicons"
     print(success_text)
     with open(log_file, 'a') as f:
         print(
@@ -253,18 +236,5 @@ def primer_blast(data_dir, db, query_path, amplicons, max_length, n_threads, log
         )
     print("\n#### off-target search finished ####\n")
 
-    return amplicons, off_target_amplicons
-
+    return amplicons
 
-def write_BLAST_warning(off_target_amplicons, amplicon_scheme, log_file):
-    """
-    for each primer pair that has potential unspecific amplicons
-    write warnings to file.
-    """
-    for amp in off_target_amplicons:
-        if amp in amplicon_scheme:
-            logging.raise_error(
-                f"{amp} could produce off-targets. No better amplicon in this area was found.",
-                log_file,
-                exit=False,
-            )

varvamp/scripts/default_config.py

@@ -4,7 +4,7 @@ This contains all varVAMP parameters.
 
 # List of all known parameters. DO NOT CHANGE!
 __all__ = [
-    'BLAST_MAX_DIFF', 'BLAST_PENALTY', 'BLAST_SETTINGS', 'BLAST_SIZE_MULTI',
+    'BLAST_MAX_DIFF', 'BLAST_SETTINGS', 'BLAST_SIZE_MULTI',
     'END_OVERLAP',
     'PCR_DNA_CONC', 'PCR_DNTP_CONC', 'PCR_DV_CONC', 'PCR_MV_CONC',
     'PRIMER_3_PENALTY', 'PRIMER_GC_END', 'PRIMER_GC_PENALTY',
@@ -74,7 +74,6 @@ BLAST_SETTINGS = {  # blast settings for query search
 }
 BLAST_MAX_DIFF = 0.5  # min percent match between primer and BLAST hit (coverage and/or mismatches)
 BLAST_SIZE_MULTI = 2  # multiplier for the max_amp size of off targets (in relation to max amp size)
-BLAST_PENALTY = 50  # amplicon penalty increase -> considered only if no other possibilities
 
 # nucleotide definitions, do NOT change
 NUCS = set("atcg")

varvamp/scripts/logging.py

@@ -15,12 +15,12 @@ from varvamp.scripts import config
 from varvamp import __version__
 
 
-def create_dir_structure(dir):
+def create_dir_structure(dir_path):
     """
     create output folders and log file
     """
     cwd = os.getcwd()
-    results_dir = os.path.join(cwd, dir)
+    results_dir = os.path.join(cwd, dir_path)
     data_dir = os.path.join(results_dir, "data/")
     # create folders
     if not os.path.exists(results_dir):
@@ -291,7 +291,6 @@ def confirm_config(args, log_file):
         (
             "BLAST_MAX_DIFF",
             "BLAST_SIZE_MULTI",
-            "BLAST_PENALTY"
         )
     ]
 
@@ -384,7 +383,6 @@ def confirm_config(args, log_file):
         ("qpcr deletion size still considered for deltaG calculation", config.QAMPLICON_DEL_CUTOFF),
         ("maximum difference between primer and blast db", config.BLAST_MAX_DIFF),
         ("multiplier of the maximum length for non-specific amplicons", config.BLAST_SIZE_MULTI),
-        ("blast penalty for off targets", config.BLAST_PENALTY)
     ]
     for var_type, var in non_negative_var:
         if var < 0:
@@ -468,11 +466,6 @@ def confirm_config(args, log_file):
             log_file,
             exit=True
         )
-    if config.BLAST_PENALTY < 10:
-        raise_error(
-            "giving a too small penalty could result in the selection of off-target producing amplicons in the final scheme.",
-            log_file,
-        )
     # confirm proper BLAST settings in dictionary
     if not isinstance(config.BLAST_SETTINGS, dict):
         raise_error(
@@ -574,7 +567,7 @@ def goodbye_message():
         "Thank you. Come again.",
         ">Placeholder for your advertisement<",
         "Make primers great again!",
-        "Ciao cacao!"
+        "Ciao cacao!",
         "And now lets pray to the PCR gods.",
         "**bibobibobop** task finished",
         "Thank you for traveling with varVAMP.",
@@ -588,5 +581,14 @@ def goodbye_message():
         "Barba non facit philosophum.",
         "Task failed successfully.",
         "Never gonna give you up, never gonna let you down.",
+        "Have you tried turning it off and on again?",
+        "Look, I am your primer scheme.",
+        "Quod erat demonstrandum.",
+        "Miau?",
+        "This is an automated message informing you that you are awsome.",
+        "Why was the negative-sense virus angry at the positive-sense virus?\nBecause he was left stranded!",
+        "If you see this message twice, you are an experienced user.",
+        "No one expects the spanish inquisition!",
+        "Primer design you must."
     ]
     print(f"\n{random.choice(messages)}")

varvamp/scripts/primers.py

@@ -386,13 +386,13 @@ def find_best_primers(left_primer_candidates, right_primer_candidates):
         primer_candidates.sort(key=lambda x: (x[3], x[1]))
         # ini everything with the primer with the lowest penalty
         to_retain = [primer_candidates[0]]
-        primer_ranges = list(range(primer_candidates[0][1], primer_candidates[0][2]+1))
+        primer_ranges = list(range(primer_candidates[0][1], primer_candidates[0][2]))
         primer_set = set(primer_ranges)
 
         for primer in primer_candidates:
             # get the thirds of the primer, only consider the middle
             thirds_len = int((primer[2] - primer[1])/3)
-            primer_positions = list(range(primer[1] + thirds_len, primer[2] - thirds_len + 1))
+            primer_positions = list(range(primer[1] + thirds_len, primer[2] - thirds_len))
             # check if none of the nucleotides of the next primer
             # are already covered by a better primer
             if not any(x in primer_positions for x in primer_set):

varvamp/scripts/qpcr.py

@@ -211,13 +211,13 @@ def assess_amplicons(left_subset, right_subset, qpcr_probes, probe, majority_con
             if "LEFT" in probe:
                 if not qpcr_probes[probe][1] in range(
                         left_primer[2] + config.QPROBE_DISTANCE[0],
-                        left_primer[2] + config.QPROBE_DISTANCE[1] + 1
+                        left_primer[2] + config.QPROBE_DISTANCE[1]
                 ):
                     continue
             elif "RIGHT" in probe:
                 if not right_primer[1] in range(
                         qpcr_probes[probe][2] + config.QPROBE_DISTANCE[0],
-                        qpcr_probes[probe][2] + config.QPROBE_DISTANCE[1] + 1
+                        qpcr_probes[probe][2] + config.QPROBE_DISTANCE[1]
 
                 ):
                     continue
@@ -258,7 +258,7 @@ def find_qcr_schemes(qpcr_probes, left_primer_candidates, right_primer_candidate
     there is no need to consider this primer probe combination.
     """
 
-    qpcr_scheme_candidates = {}
+    qpcr_scheme_candidates = []
     found_amplicons = []
     amplicon_nr = -1
 
@@ -279,15 +279,16 @@ def find_qcr_schemes(qpcr_probes, left_primer_candidates, right_primer_candidate
         # populate the primer dictionary:
         amplicon_nr += 1
         found_amplicons.append(primer_combination)
-        qpcr_scheme_candidates[f"AMPLICON_{amplicon_nr}"] = {
-            "penalty": qpcr_probes[probe][3] + primer_combination[0][3] + primer_combination[1][3],
-            "PROBE": qpcr_probes[probe],
-            "LEFT": primer_combination[0],
-            "RIGHT": primer_combination[1]
-        }
+        qpcr_scheme_candidates.append(
+            {
+                "id": f"AMPLICON_{amplicon_nr}",
+                "penalty": qpcr_probes[probe][3] + primer_combination[0][3] + primer_combination[1][3],
+                "PROBE": qpcr_probes[probe],
+                "LEFT": primer_combination[0],
+                "RIGHT": primer_combination[1]
+            }
+        )
     # and again sort by total penalty (left + right + probe)
-    qpcr_scheme_candidates = dict(sorted(qpcr_scheme_candidates.items(), key=lambda x: x[1]["penalty"]))
-
     return qpcr_scheme_candidates
 
 
@@ -296,21 +297,17 @@ def process_single_amplicon_deltaG(amplicon, majority_consensus):
     Process a single amplicon to test its deltaG and apply filtering.
     This function will be called concurrently by multiple threads.
     """
-    name, data = amplicon
-    start = data["LEFT"][1]
-    stop = data["RIGHT"][2]
-    seq = majority_consensus[start:stop]
+    seq = majority_consensus[amplicon["LEFT"][1]:amplicon["RIGHT"][2]]
     seq = seq.replace("N", "")
     seq = seq.replace("n", "")
-    amp_positions = list(range(start, stop + 1))
     # check if the amplicon overlaps with an amplicon that was previously
     # found and had a high enough deltaG
-    min_temp = min((primers.calc_temp(data["LEFT"][0]),
-                    primers.calc_temp(data["RIGHT"][0])))
+    min_temp = min((primers.calc_temp(amplicon["LEFT"][0]),
+                    primers.calc_temp(amplicon["RIGHT"][0])))
     # calculate deltaG at the minimal primer temp
-    deltaG = seqfold.dg(seq, min_temp)
+    amplicon["deltaG"] = seqfold.dg(seq, min_temp)
 
-    return deltaG, amp_positions, name
+    return amplicon
 
 
 def test_amplicon_deltaG_parallel(qpcr_schemes_candidates, majority_consensus, n_to_test, deltaG_cutoff, n_threads):
@@ -319,29 +316,34 @@ def test_amplicon_deltaG_parallel(qpcr_schemes_candidates, majority_consensus, n
     and filters if they fall below the cutoff. Multiple processes are used
     for processing amplicons in parallel.
     """
-    final_schemes = {}
-    passed_counter = 0  # counter for re-naming amplicons that passed deltaG cutoff
-    amplicon_set = set()
+    final_amplicons = []
 
     # Create a pool of processes to handle the concurrent processing
     with multiprocessing.Pool(processes=n_threads) as pool:
         # Create a list of the first n amplicon tuples for processing
-        amplicons = itertools.islice(qpcr_schemes_candidates.items(), n_to_test)
+        # The list is sorted first on whether offset targets were predicted for the amplicon,
+        # then by penalty. This ensures that amplicons with offset targets are always considered last
+        amplicons = itertools.islice(
+            sorted(qpcr_schemes_candidates, key=lambda x: (x.get("offset_targets", False), x["penalty"])),
+            n_to_test
+        )
         # process amplicons concurrently
         results = pool.starmap(process_single_amplicon_deltaG, [(amp, majority_consensus) for amp in amplicons])
         # Process the results
-        for deltaG, amp_positions, amp_name in results:
+        retained_ranges = []
+        for amp in results:
             # check if the amplicon overlaps with an amplicon that was previously
             # found and had a high enough deltaG
-            if any(x in amp_positions for x in amplicon_set):
+            if amp["deltaG"] <= deltaG_cutoff:
                 continue
-            # and if this passes cutoff make a dict entry and do not allow further
-            # amplicons in that region (they will have a lower penalty)
-            if deltaG > deltaG_cutoff:
-                new_name = f"QPCR_SCHEME_{passed_counter}"
-                final_schemes[new_name] = qpcr_schemes_candidates[amp_name]
-                final_schemes[new_name]["deltaG"] = deltaG
-                amplicon_set.update(amp_positions)
-                passed_counter += 1
-
-    return final_schemes
+            amp_range = range(amp["LEFT"][1], amp["RIGHT"][2])
+            overlaps_retained = False
+            for r in retained_ranges:
+                if amp_range.start < r.stop and r.start < amp_range.stop:
+                    overlaps_retained = True
+                    break
+            if not overlaps_retained:
+                final_amplicons.append(amp)
+                retained_ranges.append(amp_range)
+
+    return final_amplicons