Clearer code

CHANGELOG.md

@@ -1,10 +1,20 @@
+# 0.2.0dev
+
+## New features
+
+- Added the `defaults` type to the `info` and `format` configuration. Defaults for resolvable fields can be set this way.
+
 # 0.1.2 - Adjust
+
 ## Fixes
+
 - Fixed a bug where the output VCF has no header when the input VCF has no variants
 
 
 # 0.1.1 - Transform
+
 ## Fixes
+
 - Fixed a bug where the samples were missing from the header
 
 # 0.1.0 - Change

README.md

@@ -26,16 +26,16 @@ The configuration file is the core of the standardization in Svync. More informa
 
 ## Installation
 ### Mamba/Conda
-This is the preffered way of installing BedGoVcf.
+This is the preferred way of installing Svync.
 
 ```bash
-mamba install -c bioconda bedgovcf
+mamba install -c bioconda svync
 ```
 
 or with conda:
 
 ```bash 
-conda install -c bioconda bedgovcf
+conda install -c bioconda svync
 ```
 
 ### Precompiled binaries
@@ -45,17 +45,17 @@ Precompiled binaries are available for Linux and macOS on the [releases page](ht
 ### Installation from source
 Make sure you have go installed on your machine (or [install](https://go.dev/doc/install) it if you don't currently have it)
 
-Then run these commands to install bedgovcf:
+Then run these commands to install svync:
 
 ```bash
 go get .
 go build .
-sudo mv bedgovcf /usr/local/bin/
+sudo mv svync /usr/local/bin/
 ```
 
 Next run this command to check if it was correctly installed:
 
 ```bash
-bedgovcf --help
+svync --help
 ```
 

docs/configuration.md

@@ -41,6 +41,9 @@ info:
 ### value
 The `value` field can be used to change the default value of the info field. The value can be resolved (see [Resolvable fields](#resolvable-fields)).
 
+### defaults
+The `defaults` field can be used to define defaults for resolvable `INFO` and `FORMAT` fields. These defaults will be used when the required field is missing from the variant. 
+
 ### type
 The `type` field can be used to set the type of the info field (This will be reflected in the header of the output VCF file).
 
@@ -58,6 +61,8 @@ For example when all `SVLEN` info fields are positive, you maybe want to change
 info:
   SVLEN:
     value: $INFO/SVLEN
+    defaults:
+      $INFO/SVLEN: "-1"
     type: Integer
     description: "Structural variant length"
     number: 1
@@ -71,6 +76,9 @@ The `format` section can be used to change the format fields for each variant. T
 format:
   <format_field>:
     value: <new_value>
+    defaults:
+      <resolvable-field>: <default>
+      <resolvable-field>: <default>
     type: <new_type>
     description: <new_description>
     number: <new_number>

svync.go

@@ -13,7 +13,7 @@ func main() {
 		Name:            "svync",
 		Usage:           "A tool to standardize VCF files from structural variant callers",
 		HideHelpCommand: true,
-		Version:         "0.1.2",
+		Version:         "0.2.0dev",
 		Flags: []cli.Flag{
 			&cli.BoolFlag{
 				Name:     "nodate",

svync_api/config.go

@@ -79,7 +79,10 @@ func (config *Config) defineMissing() {
 	// Format fields
 	if _, ok := config.Format["GT"]; !ok {
 		config.Format["GT"] = ConfigInput{
-			Value:       "$FORMAT/GT",
+			Value: "$FORMAT/GT",
+			Defaults: map[string]string{
+				"$FORMAT/GT": "./.",
+			},
 			Number:      "1",
 			Type:        "String",
 			Description: "Genotype",

svync_api/execute.go

@@ -20,10 +20,10 @@ func Execute(Cctx *cli.Context, config *Config) {
 
 	file := Cctx.String("input")
 	inputVcf, err := os.Open(file)
-	defer inputVcf.Close()
 	if err != nil {
 		logger.Fatal(err)
 	}
+	defer inputVcf.Close()
 	header := newHeader()
 	breakEndVariants := &map[string]Variant{}
 	headerIsMade := false

svync_api/resolve.go

@@ -12,7 +12,7 @@ import (
 )
 
 // Resolve a value
-func ResolveValue(input string, variant *Variant, format *VariantFormat, Cctx *cli.Context) string {
+func ResolveValue(input string, variant *Variant, format *VariantFormat, Cctx *cli.Context, config *Config) string {
 	logger := log.New(os.Stderr, "", 0)
 
 	// Replace all the FORMAT fields
@@ -21,16 +21,19 @@ func ResolveValue(input string, variant *Variant, format *VariantFormat, Cctx *c
 	if len(allFormats) > 0 && format == nil {
 		logger.Fatalf("Cannot use a FORMAT field in a non-FORMAT context, please check your config file")
 	}
-	for _, stringToReplace := range allFormats {
-		fieldSlice := strings.Split(stringToReplace, "/")
+	for _, rawField := range allFormats {
+		fieldSlice := strings.Split(rawField, "/")
 
 		field := fieldSlice[1]
 		formatValue, ok := format.Content[field]
 
-		// TODO implement some alternative way to handle missing fields
 		if !ok {
-			if !Cctx.Bool("mute-warnings") {
-				logger.Printf("The field %s is not present in the FORMAT fields of the variant with ID %s, excluding it from this variant", field, variant.Id)
+			// Check if the field is a default value
+			defaults := config.Format[field].Defaults
+			if defaultValue, ok := defaults[rawField]; ok {
+				formatValue = []string{defaultValue}
+			} else if !Cctx.Bool("mute-warnings") {
+				logger.Printf("The field %s is not present in the FORMAT fields of the variant with ID %s, excluding it from this variant. Supply a default to mute this warning", field, variant.Id)
 			}
 		} else if len(fieldSlice) > 2 {
 			index, err := strconv.ParseInt(fieldSlice[2], 0, 64)
@@ -39,23 +42,26 @@ func ResolveValue(input string, variant *Variant, format *VariantFormat, Cctx *c
 			}
 			formatValue = []string{formatValue[index]}
 		}
-		input = strings.ReplaceAll(input, stringToReplace, strings.Join(formatValue, ","))
+		input = strings.ReplaceAll(input, rawField, strings.Join(formatValue, ","))
 	}
 
 	// Replace all the INFO fields
 	infoRegex := regexp.MustCompile(`\$INFO/[\w\d]+(/\d+)?`)
 	allInfos := infoRegex.FindAllString(input, -1)
-	for _, stringToReplace := range allInfos {
-		fieldSlice := strings.Split(stringToReplace, "/")
+	for _, rawField := range allInfos {
+		fieldSlice := strings.Split(rawField, "/")
 
 		field := fieldSlice[1]
 		info, ok := variant.Info[field]
 
-		// TODO implement some alternative way to handle missing fields
 		if !ok {
+			// Check if the field is a default value
+			defaults := config.Info[field].Defaults
 			infoType := variant.Header.Info[field].Type
-			if infoType != "Flag" && !Cctx.Bool("mute-warnings") {
-				logger.Printf("The field %s is not present in the INFO fields of the variant with ID %s, excluding it from this variant", field, variant.Id)
+			if defaultValue, ok := defaults[rawField]; ok {
+				info = []string{defaultValue}
+			} else if infoType != "Flag" && !Cctx.Bool("mute-warnings") {
+				logger.Printf("The field %s is not present in the FORMAT fields of the variant with ID %s, excluding it from this variant. Supply a default to mute this warning", field, variant.Id)
 			}
 		} else if len(fieldSlice) > 2 {
 			index, err := strconv.ParseInt(fieldSlice[2], 0, 64)
@@ -64,7 +70,7 @@ func ResolveValue(input string, variant *Variant, format *VariantFormat, Cctx *c
 			}
 			info = []string{info[index]}
 		}
-		input = strings.ReplaceAll(input, stringToReplace, strings.Join(info, ","))
+		input = strings.ReplaceAll(input, rawField, strings.Join(info, ","))
 	}
 
 	// Replace POS fields

svync_api/standardize.go

@@ -102,7 +102,7 @@ func (variant *Variant) standardize(config *Config, Cctx *cli.Context, count int
 		}
 	}
 
-	standardizedVariant.Id = fmt.Sprintf("%s_%v", ResolveValue(config.Id, variant, nil, Cctx), count)
+	standardizedVariant.Id = fmt.Sprintf("%s_%v", ResolveValue(config.Id, variant, nil, Cctx, config), count)
 
 	// Add info fields
 	for name, infoConfig := range config.Info {
@@ -114,7 +114,7 @@ func (variant *Variant) standardize(config *Config, Cctx *cli.Context, count int
 		if value == "" {
 			continue
 		}
-		standardizedVariant.Info[name] = []string{ResolveValue(value, variant, nil, Cctx)}
+		standardizedVariant.Info[name] = []string{ResolveValue(value, variant, nil, Cctx, config)}
 	}
 
 	// Add format fields
@@ -127,7 +127,7 @@ func (variant *Variant) standardize(config *Config, Cctx *cli.Context, count int
 			if val, ok := formatConfig.Alts[sVType]; ok {
 				value = val
 			}
-			newFormat.Content[name] = []string{ResolveValue(value, variant, &format, Cctx)}
+			newFormat.Content[name] = []string{ResolveValue(value, variant, &format, Cctx, config)}
 		}
 		standardizedVariant.Format[sample] = *newFormat
 	}