levseq 1.1.0__tar.gz → 1.2.1__tar.gz

{levseq-1.1.0/levseq.egg-info → levseq-1.2.1}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: levseq
-Version: 1.1.0
+Version: 1.2.1
 Home-page: https://github.com/fhalab/levseq/
 Author: Yueming Long, Emreay Gursoy, Ariane Mora, Francesca-Zhoufan Li
 Author-email: ylong@caltech.edu
@@ -48,7 +48,7 @@ Requires-Dist: tqdm
 
 In directed evolution, sequencing every variant enhances data insight and creates datasets suitable for AI/ML methods. This method is presented as an extension of the original Every Variant Sequencer using Illumina technology. With this approach, sequence variants can be generated within a day at an extremely low cost.
 
-![Figure 1: LevSeq Workflow](manuscript/Figures/LevSeq_Figure-1.png)
+![Figure 1: LevSeq Workflow](manuscript/figures/LevSeq_Figure-1.png)
 Figure 1: Overview of the LevSeq variant sequencing workflow using Nanopore technology. This diagram illustrates the key steps in the process, from sample preparation to data analysis and visualization.
 
 

{levseq-1.1.0 → levseq-1.2.1}/README.md

@@ -2,7 +2,7 @@
 
 In directed evolution, sequencing every variant enhances data insight and creates datasets suitable for AI/ML methods. This method is presented as an extension of the original Every Variant Sequencer using Illumina technology. With this approach, sequence variants can be generated within a day at an extremely low cost.
 
-![Figure 1: LevSeq Workflow](manuscript/Figures/LevSeq_Figure-1.png)
+![Figure 1: LevSeq Workflow](manuscript/figures/LevSeq_Figure-1.png)
 Figure 1: Overview of the LevSeq variant sequencing workflow using Nanopore technology. This diagram illustrates the key steps in the process, from sample preparation to data analysis and visualization.
 
 

{levseq-1.1.0 → levseq-1.2.1}/levseq/__init__.py

@@ -18,7 +18,7 @@
 __title__ = 'levseq'
 __description__ = 'LevSeq nanopore sequencing'
 __url__ = 'https://github.com/fhalab/levseq/'
-__version__ = '1.1.0'
+__version__ = '1.2.1'
 __author__ = 'Yueming Long, Emreay Gursoy, Ariane Mora, Francesca-Zhoufan Li'
 __author_email__ = 'ylong@caltech.edu'
 __license__ = 'GPL3'

{levseq-1.1.0 → levseq-1.2.1}/levseq/run_levseq.py

@@ -221,7 +221,7 @@ def demux_fastq(file_to_fastq, result_folder, barcode_path):
         executable_path = package_root / "levseq" / "barcoding" / executable_name
     if not executable_path.exists():
         raise FileNotFoundError(f"Executable not found: {executable_path}")
-    seq_min = 150 
+    seq_min = 200 
     seq_max = 10000
     prompt = f"{executable_path} -f {file_to_fastq} -d {result_folder} -b {barcode_path} -w 100 -r 100 -m {seq_min} -x {seq_max}"
     subprocess.run(prompt, shell=True, check=True)
@@ -258,7 +258,7 @@ def create_df_v(variants_df):
 
     # Translate nc_variant to aa_variant
     df_variants_["aa_variant"] = df_variants_["nc_variant"].apply(
-        lambda x: "Deletion" if x == "Deletion" else translate(x)
+    lambda x: x if x in ["Deletion", "#N.A.#"] else translate(x)
     )
     # Fill in 'Deletion' in 'aa_variant' column
     df_variants_.loc[
@@ -284,10 +284,13 @@ def create_df_v(variants_df):
     df_variants_["Alignment Probability"] = df_variants_["Alignment Probability"].fillna(0.0)
     df_variants_["Alignment Count"] = df_variants_["Alignment Count"].fillna(0.0)
 
-    # Fill in Deletion into Substitutions Column
+    # Fill in Deletion into Substitutions Column, keep #N.A.# unchanged
     for i in df_variants_.index:
         if df_variants_["nc_variant"].iloc[i] == "Deletion":
             df_variants_.Substitutions.iat[i] = df_variants_.Substitutions.iat[i].replace("", "-")
+        elif df_variants_["nc_variant"].iloc[i] == "#N.A.#":
+            df_variants_.Substitutions.iat[i] = "#N.A.#"
+
     
     # Add row and columns
     Well = df_variants_["Well"].tolist()
@@ -312,40 +315,41 @@ def create_df_v(variants_df):
     df_variants_["Plate"] = df_variants_["Plate"].apply(
         lambda x: f"0{x}" if len(x) == 1 else x
     )
-	# Rename columns as per the request
-	df_variants_.rename(columns={
-		"Variant": "nucleotide_mutation",
-		"Substitutions": "amino-acid_substitutions",
-		"nc_variant": "nt_sequence",
-		"aa_variant": "aa_sequence"
+    # Rename columns as per the request
+    df_variants_.rename(columns={
+	"Variant": "nucleotide_mutation",
+	"Substitutions": "amino-acid_substitutions",
+	"nc_variant": "nt_sequence",
+	"aa_variant": "aa_sequence"
 	}, inplace=True)
 
+
 	# Select the desired columns in the desired order
-	restructured_df = df_variants_[
-		[
-			"barcode_plate",
-			"Plate",
-			"Well",
-			"Alignment Count",
-            "nucleotide_mutation",
-            "amino-acid_substitutions",
-            "Alignment Probability",
-			"Average mutation frequency",
-			"P value",
-			"P adj. value",
-			"nt_sequence",
-			"aa_sequence",
-		]
-	]
-
-	return restructured_df, df_variants_
+    restructured_df = df_variants_[
+            [
+                    "barcode_plate",
+                    "Plate",
+                    "Well",
+                    "Alignment Count",
+                    "nucleotide_mutation",
+                    "amino-acid_substitutions",
+                    "Alignment Probability",
+                    "Average mutation frequency",
+                    "P value",
+                    "P adj. value",
+                    "nt_sequence",
+                    "aa_sequence",
+            ]
+    ]
+
+    return restructured_df, df_variants_
 
 # Helper functions for create_df_v
 def create_nc_variant(variant, refseq):
     if isinstance(variant, np.ndarray):
         variant = variant.tolist()
     if variant == "" or pd.isnull(variant):
-        return refseq
+        return "#N.A.#"  # Return #N.A.# if variant is empty or null
     elif variant == "#PARENT#":
         return refseq
     elif "DEL" in variant:
@@ -362,19 +366,25 @@ def create_nc_variant(variant, refseq):
                 nc_variant[position] = new
         return "".join(nc_variant)
 
+
 def is_valid_dna_sequence(sequence):
     return all(nucleotide in 'ATGC' for nucleotide in sequence) and len(sequence) % 3 == 0
 
 def get_mutations(row):
     try:
-        refseq = row["refseq"]
+        alignment_count = row["Alignment Count"]
         
+        # Check if alignment_count is zero and return "#N.A.#" if true
+        if alignment_count == 0:
+            return "#N.A.#"
+        
+        refseq = row["refseq"]
+
         if not is_valid_dna_sequence(refseq):
             return "Invalid refseq provided, check template sequence. Only A, T, G, C and sequence dividable by 3 are accepted."
 
         refseq_aa = translate(refseq)
         variant_aa = row["aa_variant"]
-        alignment_count = row["Alignment Count"]
 
         if variant_aa == "Deletion":
             return ""
@@ -400,6 +410,7 @@ def get_mutations(row):
         )   
         raise
 
+
 # Save plate maps and CSV
 def save_platemap_to_file(heatmaps, outputdir, name, show_msa):
     if not os.path.exists(os.path.join(outputdir, "Platemaps")):

{levseq-1.1.0 → levseq-1.2.1}/levseq/seqfit.py

@@ -124,7 +124,7 @@ def normalise_calculate_stats(processed_plate_df, value_columns, normalise='stan
         for plate in set(processed_plate_df['Plate'].values):
             for value_column in value_columns:
                 sub_df = processed_plate_df[processed_plate_df['Plate'] == plate]
-                parent_values = sub_df[sub_df['Mutations'] == parent][value_column].values
+                parent_values = sub_df[sub_df['amino-acid_substitutions'] == parent][value_column].values
                 parent_mean = np.mean(parent_values)
                 parent_sd = np.std(parent_values)
 
@@ -140,14 +140,14 @@ def normalise_calculate_stats(processed_plate_df, value_columns, normalise='stan
 
     sd_cutoff = 1.5  # The number of standard deviations we want above the parent values
     # Now for all the other mutations we want to look if they are significant, first we'll look at combinations and then individually
-    grouped_by_mutations = processed_plate_df.groupby('Mutations')
+    grouped_by_mutations = processed_plate_df.groupby('amino-acid_substitutions')
 
     rows = []
     for mutation, grp in tqdm(grouped_by_mutations):
         # Get the values and then do a ranksum test
         if mutation != parent:
             for value_column in normalised_value_columns:
-                parent_values = list(processed_plate_df[processed_plate_df['Mutations'] == parent][value_column].values)
+                parent_values = list(processed_plate_df[processed_plate_df['amino-acid_substitutions'] == parent][value_column].values)
                 parent_mean = np.mean(parent_values)
                 parent_sd = np.std(parent_values)
 
@@ -164,7 +164,7 @@ def normalise_calculate_stats(processed_plate_df, value_columns, normalise='stan
                 rows.append(
                     [value_column, mutation, len(grp), mean_vals, std_vals, median_vals, mean_vals - parent_mean, sig,
                      U1, p])
-    stats_df = pd.DataFrame(rows, columns=['value_column', 'mutation', 'number of wells with mutation', 'mean', 'std',
+    stats_df = pd.DataFrame(rows, columns=['value_column', 'amino-acid_substitutions', 'number of wells with amino-acid substitutions', 'mean', 'std',
                                            'median', 'amount greater than parent mean',
                                            f'greater than > {sd_cutoff} parent', 'man whitney U stat', 'p-value'])
     return stats_df
@@ -275,10 +275,13 @@ def work_up_lcms(
         return series
 
     df["Sample Vial Number"] = fill_vial_number(df["Sample Vial Number"].copy())
+    # Drop empty ones!
+    df = df[df["Sample Vial Number"] != 0]
     # Remove unwanted wells
     df = df[df["Sample Name"] != drop_string]
     # Get wells
-    df.insert(0, "Well", df["Sample Vial Number"].apply(lambda x: x.split("-")[-1]))
+
+    df.insert(0, "Well", df["Sample Vial Number"].apply(lambda x: str(x).split("-")[-1]))
     # Rename
     df = df.rename({"Sample Name": "Plate"}, axis="columns")
     # Create minimal DataFrame
@@ -290,7 +293,7 @@ def work_up_lcms(
         index=["Well", "Plate"], columns="Compound Name", values="Area", aggfunc="max"
     ).reset_index()
     # Get rows and columns
-    df.insert(1, "Column", df["Well"].apply(lambda x: int(x[1:])))
+    df.insert(1, "Column", df["Well"].apply(lambda x: int(x[1:]) if x[1:].isdigit() else None))
     df.insert(1, "Row", df["Well"].apply(lambda x: x[0]))
     # Set values as floats
     cols = products + substrates if substrates is not None else products
@@ -330,9 +333,10 @@ def process_plate_files(product: str, input_csv: str) -> pd.DataFrame:
     # Load the provided CSV file
     results_df = pd.read_csv(input_csv)
 
-    # Extract the required columns: Plate, Well, Mutations, and nc_variant, and remove rows with '#N.A.#' and NaN values
-    filtered_df = results_df[["Plate", "Well", "Mutations", "nc_variant", "aa_variant"]]
-    filtered_df = filtered_df[(filtered_df["Mutations"] != "#N.A.#")].dropna()
+    # Extract the required columns: Plate, Well, amino-acid_substitutionss, and nt_sequence, and remove rows with '#N.A.#' and NaN values
+    # barcode_plate	Plate	Well	Alignment Count	nucleotide_amino-acid_substitutions	amino-acid_substitutions	Alignment Probability	Average amino-acid_substitutions frequency	P value	P adj. value	nt_sequence	aa_sequence
+    filtered_df = results_df[["Plate", "Well", "amino-acid_substitutions", "nt_sequence", "aa_sequence"]]
+    filtered_df = filtered_df[(filtered_df["amino-acid_substitutions"] != "#N.A.#")].dropna()
 
     # Extract the unique entries of Plate
     unique_plates = filtered_df["Plate"].unique()
@@ -357,14 +361,14 @@ def process_plate_files(product: str, input_csv: str) -> pd.DataFrame:
                 plate_df = plate_object.df
                 plate_df["Plate"] = plate  # Add the plate identifier for reference
 
-                # Merge filtered_df with plate_df to retain Mutations and nc_variant columns
+                # Merge filtered_df with plate_df to retain amino-acid_substitutionss and nt_sequence columns
                 merged_df = pd.merge(
                     plate_df, filtered_df, on=["Plate", "Well"], how="left"
                 )
                 columns_order = (
-                    ["Plate", "Well", "Row", "Column", "Mutations"]
+                    ["Plate", "Well", "Row", "Column", "amino-acid_substitutions"]
                     + product
-                    + ["nc_variant", "aa_variant"]
+                    + ["nt_sequence", "aa_sequence"]
                 )
                 merged_df = merged_df[columns_order]
                 processed_data.append(merged_df)
@@ -374,13 +378,13 @@ def process_plate_files(product: str, input_csv: str) -> pd.DataFrame:
         processed_df = pd.concat(processed_data, ignore_index=True)
     else:
         processed_df = pd.DataFrame(
-            columns=["Plate", "Well", "Row", "Column", "Mutations"]
+            columns=["Plate", "Well", "Row", "Column", "amino-acid_substitutions"]
             + product
-            + ["nc_variant", "aa_variant"]
+            + ["nt_sequence", "aa_sequence"]
         )
 
     # Ensure all entries in 'Mutations' are treated as strings
-    processed_df["Mutations"] = processed_df["Mutations"].astype(str)
+    processed_df["amino-acid_substitutions"] = processed_df["amino-acid_substitutions"].astype(str)
 
     # Remove any rows with empty values
     processed_df = processed_df.dropna()
@@ -420,19 +424,19 @@ def match_plate2parent(df: pd.DataFrame, parent_dict: Optional[Dict] = None) ->
 
     if parent_dict is None:
 
-        # add aa_variant column if not present by translating from the nc_variant column
-        if "aa_variant" not in df.columns:
-            df["aa_variant"] = df["nc_variant"].apply(
-                Bio.sequence.Sequence(df["nc_variant"]).translate
+        # add aa_sequence column if not present by translating from the nt_sequence column
+        if "aa_sequence" not in df.columns:
+            df["aa_sequence"] = df["nt_sequence"].apply(
+                Bio.sequence.Sequence(df["nt_sequence"]).translate
             )
 
         # get all the parents from the df
-        parents = df[df["Mutations"] == "#PARENT#"].reset_index(drop=True).copy()
+        parents = df[df["amino-acid_substitutions"] == "#PARENT#"].reset_index(drop=True).copy()
 
-        # get the parent nc_variant
+        # get the parent nt_sequence
         parent_aas = (
-            df[df["Mutations"] == "#PARENT#"][["Mutations", "aa_variant"]]
-            .drop_duplicates()["aa_variant"]
+            df[df["amino-acid_substitutions"] == "#PARENT#"][["amino-acid_substitutions", "aa_sequence"]]
+            .drop_duplicates()["aa_sequence"]
             .tolist()
         )
 
@@ -440,7 +444,7 @@ def match_plate2parent(df: pd.DataFrame, parent_dict: Optional[Dict] = None) ->
 
     # get the plate names for each parent
     parent2plate = {
-        p_name: df[df["aa_variant"] == p_seq]["Plate"].unique().tolist()
+        p_name: df[df["aa_sequence"] == p_seq]["Plate"].unique().tolist()
         for p_name, p_seq in parent_dict.items()
     }
 
@@ -516,7 +520,7 @@ def norm2parent(plate_df: pd.DataFrame) -> pd.DataFrame:
 
     # get all the parents from the df
     parents = (
-        plate_df[plate_df["Mutations"] == "#PARENT#"].reset_index(drop=True).copy()
+        plate_df[plate_df["amino-acid_substitutions"] == "#PARENT#"].reset_index(drop=True).copy()
     )
     filtered_parents = (
         parents.drop(index=detect_outliers_iqr(parents["pdt"]))
@@ -610,7 +614,7 @@ def get_single_ssm_site_df(
     # get parents from those plates
     site_parent_df = (
         single_ssm_df[
-            (single_ssm_df["Mutations"] == "#PARENT#")
+            (single_ssm_df["amino-acid_substitutions"] == "#PARENT#")
             & (single_ssm_df["Plate"].isin(site_df["Plate"].unique()))
         ]
         .reset_index(drop=True)
@@ -1145,12 +1149,12 @@ def gen_seqfitvis(
 
     df = pd.read_csv(seqfit_path)
     # ignore deletion meaning "Mutations" == "-"
-    df = df[df["Mutations"] != "-"].copy()
+    df = df[df["amino-acid_substitutions"] != "-"].copy()
     # count number of sites mutated and append mutation details
     # df["num_sites"] = df['Mutations'].apply(lambda x: 0 if x == "#PARENT#" else len(x.split("_")))
 
     # Apply function to the column
-    df[["num_sites", "mut_dets"]] = df["Mutations"].apply(process_mutation)
+    df[["num_sites", "mut_dets"]] = df["amino-acid_substitutions"].apply(process_mutation)
 
     # apply the norm function to all plates
     df = df.groupby("Plate").apply(norm2parent).reset_index(drop=True).copy()

{levseq-1.1.0 → levseq-1.2.1}/levseq/visualization.py

@@ -142,7 +142,7 @@ def _make_platemap(df, title, cmap=None):
             'Row': list('ABCDEFGH'),
             'Column': [str(i) for i in range(1, 13)],
             'logseqdepth': [0] * 96,
-            'Substitutions': [''] * 96,
+            'amino-acid_substitutions': [''] * 96,
             'Alignment Count': [0] * 96,
             'Alignment Probability': [0] * 96
         })
@@ -195,7 +195,7 @@ def _make_platemap(df, title, cmap=None):
 
     # add tooltips
     tooltips = [
-        ("Substitutions", "@Substitutions"),
+        ("amino-acid_substitutions", "@amino-acid_substitutions"),
         ("Alignment Count", "@Alignment Count"),
         ("Alignment Probability", "@Alignment Probability"),
     ]
@@ -211,7 +211,7 @@ def _make_platemap(df, title, cmap=None):
             kdims=["Column", "Row"],
             vdims=[
                 "logseqdepth",
-                "Substitutions",
+                "amino-acid_substitutions",
                 "Alignment Count",
                 "Alignment Probability",
             ],
@@ -300,7 +300,7 @@ def _make_platemap(df, title, cmap=None):
         return new_line_mutations
 
     _df = df.copy()
-    _df["Labels"] = _df["Substitutions"].apply(split_variant_labels)
+    _df["Labels"] = _df["amino-acid_substitutions"].apply(split_variant_labels)
 
     # Set the font size based on if #PARENT# is in a well and num of mutations
     max_num_mutations = _df["Labels"].apply(lambda x: len(x.split("\n"))).max()

{levseq-1.1.0 → levseq-1.2.1/levseq.egg-info}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: levseq
-Version: 1.1.0
+Version: 1.2.1
 Home-page: https://github.com/fhalab/levseq/
 Author: Yueming Long, Emreay Gursoy, Ariane Mora, Francesca-Zhoufan Li
 Author-email: ylong@caltech.edu
@@ -48,7 +48,7 @@ Requires-Dist: tqdm
 
 In directed evolution, sequencing every variant enhances data insight and creates datasets suitable for AI/ML methods. This method is presented as an extension of the original Every Variant Sequencer using Illumina technology. With this approach, sequence variants can be generated within a day at an extremely low cost.
 
-![Figure 1: LevSeq Workflow](manuscript/Figures/LevSeq_Figure-1.png)
+![Figure 1: LevSeq Workflow](manuscript/figures/LevSeq_Figure-1.png)
 Figure 1: Overview of the LevSeq variant sequencing workflow using Nanopore technology. This diagram illustrates the key steps in the process, from sample preparation to data analysis and visualization.