Switched to JSON schema for the data configuration. So much bloated c…

…ode that is going the way of the dodo :-)

MANIFEST.in

@@ -1,5 +1,5 @@
 include DESCRIPTION.rst
-include *py
-recursive-inlcude mikado_lib *py
+include *py *.json *.yaml
+recursive-include mikado_lib *.py *.json *.yaml
 recursive-include util *py
 recursive-include test *py

mikado_lib/__init__.py

@@ -10,5 +10,5 @@
 import mikado_lib.loci_objects
 import mikado_lib.serializers
 import mikado_lib.exceptions
-import mikado_lib.json_utils
+import mikado_lib.configuration.json_utils
 import mikado_lib.scales

mikado_lib/configuration/__init__.py

@@ -0,0 +1 @@
+__author__ = 'venturil'

mikado_lib/configuration/configuration_blueprint.json

@@ -0,0 +1,233 @@
+{
+  "type": "object",
+  "comment": [""],
+  "properties": {
+    "chimera_split":{
+      "comment": ["Parameters related to the splitting of transcripts in the presence of",
+      "two or more ORFs. Parameters:",
+        "- execute: whether to split multi-ORF transcripts at all. Boolean.",
+        "- blast_check: whether to use BLAST information to take a decision. See blast_params for details.",
+        "- blast_params: Parameters related to which BLAST data we want to analyse."],
+      "type": "object",
+      "properties":{
+        "blast_check": {"type": "boolean", "default": false},
+        "execute": {"type": "boolean", "default": false},
+        "blast_params": {
+          "type": "object",
+          "comment": ["Parameters for the BLAST check prior to splitting.",
+            "- evalue: Minimum evalue for the whole hit. Default: 1e-6",
+            "- hsp_evalue: Minimum evalue for any HSP hit (some might be discarded even if the whole hit is valid). Default: 1e-6",
+            "- leniency: One of 'STRINGENT', 'LENIENT', 'PERMISSIVE'. Default: LENIENT",
+            "- max_target_seqs: maximum number of hits to consider. Default: 3",
+            "- minimal_hsp_overlap: minimum overlap of the ORF with the HSP (*not* reciprocal). Default: 0.9, i.e. 90%",
+            "- min_overlap_duplication: minimum overlap (in %) for two ORFs to consider them as target duplications.",
+            "  This means that if two ORFs have no HSPs in common, but the coverage of their disjoint HSPs covers more",
+            "  Than this % of the length of the *target*, they represent most probably a duplicated gene."
+          ],
+          "properties": {
+             "evalue": {"type": "number", "minimum": 0, "default": 0.000001},
+             "hsp_evalue": {"type": "number", "minimum": 0, "default": 0.000001},
+             "leniency": {"type": "string", "enum": ["STRINGENT", "LENIENT","PERMISSIVE"],
+             "default": "LENIENT"},
+             "max_target_seqs": {"type": "integer", "minimum": 1, "default": 3},
+             "minimal_hsp_overlap": {"type": "number", "minimum": 0,
+               "maximum": 1, "default": 0.9},
+            "min_overlap_duplication": {"type": "number", "minimum": 0, "maximum": 1, "default": 0.9}
+          }
+        }
+      },
+      "required": ["blast_check", "execute"]
+    },
+   "log_settings":{
+     "comment": ["Settings related to the logs. Keys:",
+       "log: the name of the log file. Default: mikado.log",
+     "log_level: verbosity. In decreasing order: 'DEBUG', 'INFO', 'WARNING', 'ERROR', 'CRITICAL'"],
+     "type": "object",
+     "properties":{
+       "log": {
+         "type": "string",
+         "default": "mikado.log"
+       },
+       "log_level": {"type": "string", "enum": ["DEBUG", "INFO", "WARNING", "ERROR", "CRITICAL"],
+         "default": "WARNING" }
+     }
+   },
+  "db_settings": {
+    "comment": ["Settings related to DB connection. Parameters:",
+    "db: the DB to connect to. Required. Default: mikado.db",
+    "dbtype: Type of DB to use. Choices: sqlite, postgresql, mysql. Default: sqlite.",
+    "dbhost: Host of the database. Unused if dbtype is sqlite. Default: localhost",
+    "dbuser: DB user. Default: ''", "dbpasswd: DB password for the user. Default: ''",
+    "dbport: Integer. It indicates the default port for the DB."],
+    "type": "object",
+    "properties": {
+      "db": {"default": "mikado.db", "type": "string"},
+      "dbtype": {"default": "sqlite", "enum": ["mysql", "postgresql", "sqlite"], "type": "string" },
+      "dbhost": { "format": "hostname", "default": "localhost"},
+      "dbuser": {"type": "string", "default": ""},
+      "dbpasswd": {"type": "string", "default": ""},
+      "dbport": {"type": "integer", "default": 0}
+    },
+    "required": ["dbtype", "db"]
+  },
+  "source": { "type": "string", "default": "Mikado"},
+  "alternative_splicing": {
+    "comment": ["Parameters related to alternative splicing reporting.",
+    "- report: whether to report at all or not the AS events.",
+    "- min_cds_overlap: minimum overlap between the CDS of the primary transcript and any AS event. Default: 60%.",
+    "- keep_retained_introns: Whether to consider as valid AS events where one intron",
+      "is retained compared to the primary or any other valid AS. Default: false.",
+    "- max_isoforms: Maximum number of isoforms per locus. 1 implies no AS reported. Default: 3",
+    "- valid_ccodes: Valid class codes for AS events. See documentation for details. Choices:",
+    "   j, n, O, e, K, o, h. Default: j, O, n, h",
+    "- max_utr_length: Maximum length of the UTR for AS events. Default: 10e6 (i.e. no limit)",
+    "- max_fiveutr_length: Maximum length of the 5'UTR for AS events. Default: 10e6 (i.e. no limit)",
+    "- max_threeutr_length: Maximum length of the 5'UTR for AS events. Default: 10e6 (i.e. no limit)"],
+    "type": "object",
+    "properties": {
+      "report": {
+        "type": "boolean",
+        "default": true
+      },
+      "min_cds_overlap": {
+        "type": "number",
+        "minimum": 0,
+        "maximum": 1,
+        "default": 0.6
+      },
+      "keep_retained_introns": {
+        "type": "boolean",
+        "default": false
+      },
+      "max_isoforms": {
+        "type": "integer",
+        "minimum": 1,
+        "default": 3
+      },
+      "max_utr_length":{
+        "type": "integer",
+        "default": 1000000,
+        "minimum": 1
+      },
+      "max_fiveutr_length":{
+        "type": "integer",
+        "default": 1000000,
+        "minimum": 1
+      },
+      "max_threeutr_length":{
+        "type": "integer",
+        "default": 1000000,
+        "minimum": 1
+      },
+      "valid_ccodes": {
+        "type": "array",
+        "items": {
+          "type": "string",
+          "enum": [
+            "j",
+            "n",
+            "O",
+            "e",
+            "K",
+            "o",
+            "h"
+          ]
+        },
+        "default": [
+          "j",
+          "n",
+          "O",
+          "h"
+        ]
+      }
+    }
+  },
+    "orf_loading": {
+      "type": "object",
+      "comment": ["Parameters related to ORF loading.",
+        "- minimal_secondary_orf_length: Minimum length of a *secondary* ORF",
+        "  to be loaded after the first, in bp. Default: 200 bps",
+        "- minimal_orf_length: Minimum length in bps of an ORF to be loaded,",
+        "  as the primary ORF, onto a transcript. Default: 50 bps",
+        "- strand_specific: Boolean flag. If set to true, monoexonic transcripts",
+        "  will not have their ORF reversed even if they would have an ORF on the opposite strand."
+      ],
+      "properties": {
+        "minimal_secondary_orf_length": {
+          "type": "integer",
+          "default": 200,
+          "minimum": 0
+        },
+        "minimal_orf_length": {
+          "type": "integer",
+          "default": 50,
+          "minimum": 0
+        },
+        "strand_specific": {
+          "type": "boolean",
+          "default": true
+        }
+      }
+    },
+    "run_options": {
+      "comment": ["Generic run options.",
+      "- shm: boolean flag. If set and the DB is sqlite, it will be copied onto the /dev/shm faux partition",
+      "- shm_db: String. It indicates a DB that has to be copied onto SHM and left there for",
+      "  concurrent Mikado runs.",
+      "- shm_shared: boolean flag. If set, the database loaded onto SHM will be shared and should not be",
+        "  deleted at the end of the run (see shm_db).",
+      "  for faster access. Default: false",
+      "- exclude_cds: boolean flag. If set, the CDS information will not be printed in Mikado output. Default: false",
+      "- purge: boolean flag. If set, all loci where all transcripts have a score of 0 will be excluded",
+      "  from the output. Default: false",
+      "- remove_overlapping_fragments: boolean flag. If set, fragments (defined as monoexonic loci",
+      "  classified as P,x,i or p compared to another locus, will be removed from the output.",
+      "- fragments_maximal_cds: a monoexonic superlocus will never be considered a fragment if its longest CDS is over",
+      "  this length. Default: 100 bps.",
+      "- threads: number of threads to use. Default: 1",
+      "- preload: boolean flag. If set, the whole database will be preloaded into memory for faster access. Useful when",
+        "  using SQLite databases."],
+      "type": "object",
+      "properties": {
+        "shm": {
+          "type": "boolean",
+          "default": false
+        },
+        "shm_db":{"type": "string", "default": ""},
+        "exclude_cds": {
+          "type": "boolean",
+          "default": false
+        },
+        "purge": {
+          "type": "boolean",
+          "default": false
+        },
+        "remove_overlapping_fragments": {
+          "type": "boolean",
+          "default": true
+        },
+        "fragments_maximal_cds": {
+          "type": "integer",
+          "default": 100
+        },
+        "subloci_from_cds_only": {
+          "type": "boolean",
+          "default": false
+        },
+        "threads": {
+          "type": "integer",
+          "default": 1,
+          "minimum": 1
+        },
+        "preload": {
+          "type": "boolean",
+          "default": false
+        }
+      }
+    },
+    "scoring_file": {
+      "type": "string",
+      "default": "scoring.yaml"
+    }
+  }
+}