fix: output file names

98a05c21 · Totaro Massimo G · 4b4161b5 · 98a05c21
Commit 98a05c21 authored 3 months ago by Totaro Massimo G
--- a/ssn.ipynb
+++ b/ssn.ipynb
 {
-  "nbformat": 4,
-  "nbformat_minor": 0,
-  "metadata": {
-    "colab": {
-      "provenance": []
-    },
-    "kernelspec": {
-      "name": "python3",
-      "display_name": "Python 3"
-    },
-    "language_info": {
-      "name": "python"
-    }
-  },
  "cells": [
    {
      "cell_type": "markdown",
+      "metadata": {
+        "id": "AkmFcUlSqjvW"
+      },
      "source": [
        "# SSN analysis\n",
        "\n",
@@ -27,13 +16,16 @@
        "\n",
        "Alternatively, a BLAST-preanalysed database can be provided.\n",
        "Just hit cancel on the first FASTA upload prompt and then provide the TSV file."
-      ],
-      "metadata": {
-        "id": "AkmFcUlSqjvW"
-      }
+      ]
    },
    {
      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "cellView": "form",
+        "id": "ad3vrSFIVZNl"
+      },
+      "outputs": [],
      "source": [
        "#@title Click below to run\n",
        "\n",
@@ -69,8 +61,8 @@
        "    print('No file provided')\n",
        "\n",
        "if isfile(fileTsv):\n",
-        "  fileBOut = 'evs.csv'\n",
-        "  fileEOut = 'bts.csv'\n",
+        "  fileBOut = 'bts.csv'\n",
+        "  fileEOut = 'ets.csv'\n",
        "  fileZip = NamedTemporaryFile(suffix='.zip').name\n",
        "\n",
        "  try:\n",
@@ -119,13 +111,21 @@
        "    files.download(fileZip)\n",
        "  except:\n",
        "    print('Invalid file format')"
+      ]
+    }
  ],
  "metadata": {
-        "id": "ad3vrSFIVZNl",
-        "cellView": "form"
+    "colab": {
+      "provenance": []
    },
-      "execution_count": null,
-      "outputs": []
+    "kernelspec": {
+      "display_name": "Python 3",
+      "name": "python3"
+    },
+    "language_info": {
+      "name": "python"
    }
-  ]
+  },
+  "nbformat": 4,
+  "nbformat_minor": 0
 }
 %% Cell type:markdown id: tags:

 # SSN analysis

 This is an automated script to generate a Sequence Similarity Network starting from a curated sequence alignment.

 To run the analysis, click on the button in the cell below.

 You will be prompted to upload a FASTA file containing the curated sequence alignment.

 Alternatively, a BLAST-preanalysed database can be provided.
 Just hit cancel on the first FASTA upload prompt and then provide the TSV file.

 %% Cell type:code id: tags:

 ``` 
 #@title Click below to run

 import numpy
 from google.colab import files
 from pandas import read_csv
 from zipfile import ZipFile
 from os.path import isfile
 from tempfile import NamedTemporaryFile

 fileFasta = NamedTemporaryFile(suffix='.fasta').name
 fileTsv = NamedTemporaryFile(suffix='.tsv').name

 try:
  print('Upload FASTA file to be analysed')
  files.upload_file(fileFasta)

  print('Analysing FASTA file...')

  !apt-get install ncbi-blast+ > /dev/null && \
  makeblastdb -in {fileFasta} -dbtype prot -parse_seqids -out 'DB' > /dev/null && \
  blastp -db 'DB' -query {fileFasta} -out {fileTsv} -outfmt "6 qseqid sseqid evalue bitscore"

  if not isfile(fileTsv):
    print('The provided FASTA file could not be processed')
    raise ValueError

 except ValueError:
  try:
    print('Upload TSV file to be analysed')
    files.upload_file(fileTsv)
  except ValueError:
    print('No file provided')

 if isfile(fileTsv):
-  fileBOut = 'evs.csv'
-  fileEOut = 'bts.csv'
+  fileBOut = 'bts.csv'
+  fileEOut = 'ets.csv'
  fileZip = NamedTemporaryFile(suffix='.zip').name

  try:
    df = read_csv(fileTsv,
                  sep='\t',
                  header=None,
                  index_col=[0, 1],
                  names=['T', 'S', 'e', 'b'],
                  dtype={'e':numpy.float64,
                          'b':numpy.float64})

    e = df.pivot_table(index='T',
                      columns='S',
                      values='e',
                      aggfunc='min')
    ((numpy.minimum(e, e.T) + numpy.tril(numpy.full(e.shape, numpy.nan)))
      .unstack()
      .to_frame()
      .sort_index(level=1)
      .dropna()
      .reset_index()
      .to_csv(fileEOut,
              sep=';',
              index=False,
              header=['Source','Target','evalue']))

    b = df.pivot_table(index='T',
                      columns='S',
                      values='b',
                      aggfunc='max')
    ((numpy.maximum(b, b.T) + numpy.tril(numpy.full(b.shape, numpy.nan)))
      .unstack()
      .to_frame()
      .sort_index(level=1)
      .dropna()
      .reset_index()
      .to_csv(fileBOut,
              sep=';',
              index=False,
              header=['Source','Target','bitscore']))

    with ZipFile(fileZip, 'w') as f:
      f.write(fileBOut)
      f.write(fileEOut)

    files.download(fileZip)
  except:
    print('Invalid file format')
 ```