Fix GFF parser (bebop#302)

* fixed bebop#152
---------

Co-authored-by: Timothy Stiles <[email protected]>

.gitignore

@@ -175,3 +175,4 @@ Network Trash Folder
 Temporary Items
 .apdisk
 
+.vscode/launch.json

.vscode/settings.json

@@ -0,0 +1,3 @@
+{
+	"cSpell.words": ["bioinformatics", "genbank", "polyjson"]
+}

data/ecoli-mg1655.gff

@@ -1,4 +1,7 @@
 ##gff-version 3
+#!genome-build ASM584v2 
+#!genome-build-accession NCBI_Assembly:GCA_000005845.2
+#!annotation-source NCBI RefSeq 
 ##sequence-region U00096.3 1 4641652
 U00096.3	feature	gene	190	255	.	+	.	db_xref=EcoGene:EG11277;gene=thrL;gene_synonym=ECK0001,JW4367;locus_tag=b0001
 U00096.3	feature	CDS	190	255	.	+	0	codon_start=1;db_xref=GI:1786182,ASAP:ABE-0000006,UniProtKB/Swiss-Prot:P0AD86,EcoGene:EG11277;function=leader%3B Amino acid biosynthesis: Threonine,1.5.1.8 metabolism%3B building block biosynthesis%3B amino acids%3B threonine;gene=thrL;gene_synonym=ECK0001,JW4367;locus_tag=b0001;note=GO_process: GO:0009088 - threonine biosynthetic process;product=thr operon leader peptide;protein_id=AAC73112.1;transl_table=11;translation=MKRISTTITTTITITTGNGAG

io/gff/gff.go

@@ -13,6 +13,7 @@ package gff
 
 import (
 	"bytes"
+	"errors"
 	"io"
 	"os"
 	"sort"
@@ -127,10 +128,12 @@ func Parse(file io.Reader) (Gff, error) {
 	gff.Meta.CheckSum = blake3.Sum256(fileBytes)
 
 	lines := strings.Split(gffString, "\n")
-	metaString := lines[0:2]
+	regionStringArray, endOfMetaInfo, err := extractInfoFromField(lines, "##sequence-region")
+	metaString := lines[0:endOfMetaInfo]
 	versionString := metaString[0]
-	regionStringArray := strings.Split(metaString[1], " ")
-
+	if err != nil {
+		return Gff{}, err
+	}
 	// get name for general meta
 	meta := Meta{}
 	meta.Name = regionStringArray[1] // Formally region name, but changed to name here for generality/interoperability.
@@ -154,7 +157,7 @@ func Parse(file io.Reader) (Gff, error) {
 			fastaFlag = true
 		} else if len(line) == 0 {
 			continue
-		} else if line[0:2] == "##" {
+		} else if line[0:2] == "##" || line[0:2] == "#!" {
 			continue
 		} else if fastaFlag && line[0:1] != ">" {
 			// sequence.Sequence = sequence.Sequence + line
@@ -206,6 +209,29 @@ func Parse(file io.Reader) (Gff, error) {
 	return gff, err
 }
 
+// regionString takes in the lines array,fieldName that is needed in gff file, and
+// returns the region containing fieldName if found
+// throws error if not found
+func extractInfoFromField(lines []string, fieldName string) ([]string, int, error) {
+	index := 0
+	endOfMetaInfo := 0
+	for lineIndex, line := range lines {
+		if strings.Contains(line, "#") {
+			if strings.Contains(line, fieldName) {
+				index = lineIndex
+			}
+			continue
+		}
+		endOfMetaInfo = lineIndex
+		break
+
+	}
+	if index == 0 && fieldName != "gff-version" {
+		return nil, 0, errors.New("the given file does not have any meta information")
+	}
+	return strings.Split(lines[index], " "), endOfMetaInfo, nil
+}
+
 // Build takes an Annotated sequence and returns a byte array representing a gff to be written out.
 func Build(sequence Gff) ([]byte, error) {
 	var gffBuffer bytes.Buffer