tabtk 1.0.2__tar.gz → 1.0.4__tar.gz

{tabtk-1.0.2 → tabtk-1.0.4}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: tabtk
-Version: 1.0.2
+Version: 1.0.4
 Summary: tabtk: A toolkit for sequence similarity analysis based on tabular data
 Author-email: Chrispin Chaguza <chrispin.chaguza@gmail.com>
 Maintainer-email: Chrispin Chaguza <chrispin.chaguza@gmail.com>

{tabtk-1.0.2 → tabtk-1.0.4}/tabtk/__init__.py

@@ -1,4 +1,4 @@
 """tabkit: A toolkit for sequence similarity analysis based on tabular data"""
 
-__version__ = "1.0.2"
+__version__ = "1.0.4"
 

{tabtk-1.0.2 → tabtk-1.0.4}/tabtk/tabtk.py

@@ -181,17 +181,19 @@ def blastSimilarity(cmdValues):
     blastData = pd.read_csv(tmpOut2,delimiter="\t",dtype={'pident': float,'qcovs': float})
     blastData["tmp"] = blastData.apply(lambda x: (x["qcovs"] * x["pident"])/100,axis=1)
 
+    blastDataR = blastData[["qseqid","sseqid","tmp"]].groupby(['qseqid','sseqid']).agg({'tmp':'mean'})
+    blastDataR = blastDataR.reset_index()
+
     blastData["qseqidR"] = blastData.apply(lambda x: x["qseqid"] if x["qseqid"] > x["sseqid"] else x["sseqid"],axis=1)
-    blastData["sseqidR"] = blastData.apply(lambda x: x["qseqid"] if x["qseqid"] < x["sseqid"] else x["sseqid"],axis=1)
+    blastData["sseqidR"] = blastData.apply(lambda x: x["sseqid"] if x["qseqid"] > x["sseqid"] else x["qseqid"],axis=1)
 
     blastData["qseqid"] = blastData["qseqidR"]
     blastData["sseqid"] = blastData["sseqidR"]
 
-    blastData["cov"] = blastData[["qseqid","sseqid","tmp"]].groupby(["qseqid","sseqid"])["tmp"].transform(lambda x: statistics.mean(x))
-    blastDataR = blastData.groupby(["qseqid","sseqid"]).first().reset_index()
+    blastDataR = blastData[["qseqid","sseqid","tmp"]].groupby(['qseqid','sseqid']).agg({'tmp':'min'})
+    blastDataR = blastDataR.reset_index()
 
-    blastData = blastDataR.drop(columns=["tmp","length","pident","qseqidR","sseqidR"])
-    blastDataFinal = blastData[["sseqid","qseqid","cov"]].rename(columns={"sseqid":"seq1","qseqid":"seq2","cov":"pid"}).values.tolist()
+    blastDataFinal = blastDataR[["sseqid","qseqid","tmp"]].rename(columns={"sseqid":"seq1","qseqid":"seq2","tmp":"pid"}).values.tolist()
 
     bar = ChargingBar("Cleaning temporary files", max=len(fileNames))
     for fileName in tmpFileNames:
@@ -237,18 +239,28 @@ def fastaniSimilarity(cmdValues):
 
     tmpOut = f"mash.{str(datetime.datetime.now()).replace(" ","").replace("-",".").replace(":",".")}"
 
-    fastaniCmd = f"fastANI --rl {cmdValues["data"]} --ql {cmdValues["data"]} -k {cmdValues["length"]} --matrix --fragLen {cmdValues["fragment"]} --minFraction {cmdValues["fraction"]} --output {tmpOut} -t {cmdValues["threads"]}"
+    fastaniCmd = f"fastANI --rl {cmdValues["data"]} --ql {cmdValues["data"]} --kmer {cmdValues["length"]} --matrix --fragLen {cmdValues["fragment"]} --minFraction {cmdValues["fraction"]} --output {tmpOut} --threads {cmdValues["threads"]}"
 
     print("Calculating percent similarity... may take longer... be patient...")
 
     subprocess.call(fastaniCmd,shell=True,stdout=subprocess.DEVNULL,stderr=subprocess.STDOUT)
 
     fastaniResults = []
-
+    
     with open(tmpOut,"r") as fastaniFile:
         for itemPos,fastaniOutput in enumerate(fastaniFile):
             item = fastaniOutput.strip().split("\t")
-            fastaniResults.append(f"{Path(item[0]).stem}\t{Path(item[1]).stem}\t{float(item[2])}")
+            fastaniResults.append([Path(item[0]).stem,Path(item[1]).stem,float(item[2])])
+
+    fastaniData = pd.DataFrame(fastaniResults, columns=["seq1R","seq2R","pidR"])
+    fastaniData['pidR'] = fastaniData['pidR'].astype(float)
+
+    fastaniData["seq1RR"] = fastaniData.apply(lambda x: x["seq1R"] if x["seq1R"] > x["seq2R"] else x["seq2R"],axis=1)
+    fastaniData["seq2RR"] = fastaniData.apply(lambda x: x["seq2R"] if x["seq1R"] > x["seq2R"] else x["seq1R"],axis=1)
+
+    fastaniData = fastaniData[["seq1RR","seq2RR","pidR"]].groupby(['seq1RR','seq2RR']).agg({'pidR':min})
+    fastaniDataR = fastaniData.reset_index()
+    fastaniDataFinal = fastaniDataR[["seq1RR","seq2RR","pidR"]].rename(columns={"seq1RR":"seq1","seq2RR":"seq2","pid":"pidR"}).values.tolist()
 
     if os.path.exists(tmpOut):
         os.remove(tmpOut)
@@ -260,7 +272,7 @@ def fastaniSimilarity(cmdValues):
     else:
         pass
 
-    return(fastaniResults)
+    return(fastaniDataFinal)
 
 
 def similarity(cmdValues):
@@ -314,7 +326,7 @@ def similarity(cmdValues):
 
             bar = ChargingBar('Writing similarity values', max=len(similarityResults))
             for itemPair in similarityResults:
-                simData.write(f"{itemPair}\n")
+                simData.write(f"{"\t".join([str(r) for r in itemPair])}\n")
 
                 bar.next()
             bar.finish()
@@ -1564,8 +1576,8 @@ def tabtkMain():
                  'output': options.output[0:][0] if isinstance(options.output,list) else options.output,
                  'length': int(options.length[0:][0]) if isinstance(options.length,list) else int(options.length),
                  'fragment': int(options.fragment[0:][0]) if isinstance(options.fragment,list) else int(options.fragment),
-                 'fraction': int(options.fraction[0:][0]) if isinstance(options.fraction,list) else int(options.fraction),
-                 'size': int(options.size[0:][0]) if isinstance(options.size,list) else int(options.size),
+                 'fraction': float(options.fraction[0:][0]) if isinstance(options.fraction,list) else float(options.fraction),
+                 'size': float(options.size[0:][0]) if isinstance(options.size,list) else float(options.size),
                  'mash': options.mash,
                  'fastani': options.fastani,
                  'blast': options.blast,

{tabtk-1.0.2 → tabtk-1.0.4}/tabtk.egg-info/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: tabtk
-Version: 1.0.2
+Version: 1.0.4
 Summary: tabtk: A toolkit for sequence similarity analysis based on tabular data
 Author-email: Chrispin Chaguza <chrispin.chaguza@gmail.com>
 Maintainer-email: Chrispin Chaguza <chrispin.chaguza@gmail.com>