Added spots counts and spots intensities steps

CHANGELOG.md

@@ -1,5 +1,7 @@
 # JaneliaSciComp/easifish: Changelog
 
+* Added spots count and spots intensities
+
 * Added spots warping using Bigstream
 
 * Added RS-FISH module for spot extraction

conf/modules.config

@@ -16,3 +16,5 @@ includeConfig './multiscale-modules.config'
 includeConfig './segmentation-modules.config'
 includeConfig './spot-extraction-modules.config'
 includeConfig './warp-spots-modules.config'
+includeConfig './spots-features-modules.config'
+

conf/spots-features-modules.config

@@ -0,0 +1,10 @@
+params {
+    spots_counts_cores  = 1
+    spots_counts_mem_gb = 2
+    spots_props_cores   = 1
+    spots_props_mem_gb  = 2
+    spots_counts_subdir = 'spots-counts'
+    spots_props_subdir  = 'spots-props'
+    dapi_channel        = ''
+    bleeding_channel    = ''
+}

examples/tiny_sample_params.json

@@ -33,7 +33,8 @@
 
     "cellpose_test": false,
     "cellpose_verbose": true,
-    "cellpose_dask_workers": 3,
+    "cellpose_dask_workers": 1,
+    "cellpose_dask_worker_cpus": 4,
     "cellpose_merge_with_iou_only": false,
     "cellpose_worker_runtime_opts": "--nv",
 
@@ -50,6 +51,9 @@
     "multiscale_spark_worker_cores": 20,
     "multiscale_spark_gb_per_core": 5,
 
+    "dapi_channel": "c1",
+    "bleeding_channel": "c0",
+
     "lsf_opts": "-P scicompsoft",
     "cellpose_worker_lsf_opts": "-q gpu_l4 -gpu 'num=1'"
 }

modules/local/post_rs_fish/main.nf

@@ -49,5 +49,4 @@ process POST_RS_FISH {
         post-rs-fish: v1
     END_VERSIONS
     """
-
 }

modules/local/spots/counts/main.nf

@@ -0,0 +1,60 @@
+process SPOTS_COUNTS {
+    tag { meta.id }
+    container { task.ext.container ?: 'ghcr.io/janeliascicomp/easifish-spots-utils:v1' }
+    cpus { ncpus }
+    memory { "${mem_in_gb}GB" }
+
+    input:
+    tuple val(meta),
+          path(input_image_path, stageAs: 'image/*'),
+          val(input_dataset),
+          path(labels_path, stageAs: 'labels/*'),
+          val(labels_dataset),
+          path(spots_input_dir, stageAs: 'spots/*'),
+          val(input_pattern),
+          path(output_dir, stageAs: 'output/*')
+    val(ncpus)
+    val(mem_in_gb)
+
+    output:
+    tuple val(meta),
+          env(full_input_image_path),
+          val(input_dataset),
+          env(full_spots_input_dir),
+          env(output_csv_file),      emit: results
+    path "versions.yml"            , emit: versions
+
+    when:
+    task.ext.when == null || task.ext.when
+
+    script:
+    """
+    full_input_image_path=\$(readlink -e ${input_image_path})
+    full_labels_path=\$(readlink -e ${labels_path})
+    full_spots_input_dir=\$(readlink -e ${spots_input_dir})
+    full_output_dir=\$(readlink ${output_dir})
+
+    mkdir -p \${full_output_dir}
+
+    output_csv_file="\${full_output_dir}/count.csv"
+
+    CMD=(
+        python
+        /opt/scripts/spots-utils/labeled-spots-counts.py
+        --image-container \${full_input_image_path}
+        --image-subpath ${input_dataset}
+        --labels-container \${full_labels_path}
+        --labels-subpath ${labels_dataset}
+        --spots-pattern \"\${full_spots_input_dir}/${input_pattern}\"
+        --output \${output_csv_file}
+    )
+    echo "CMD: \${CMD[@]}"
+    (exec "\${CMD[@]}")
+
+    cat <<-END_VERSIONS > versions.yml
+    "${task.process}":
+        assign-spots: v1
+    END_VERSIONS
+    """
+
+}

modules/local/spots/props/main.nf

@@ -0,0 +1,62 @@
+process SPOTS_PROPS {
+    tag { meta.id }
+    container { task.ext.container ?: 'ghcr.io/janeliascicomp/easifish-spots-utils:v1' }
+    cpus { ncpus }
+    memory { "${mem_in_gb}GB" }
+
+    input:
+    tuple val(meta),
+          path(input_image_path, stageAs: 'image/*'),
+          val(input_dataset),
+          path(labels_path, stageAs: 'labels/*'),
+          val(labels_dataset),
+          val(dapi_dataset),
+          val(bleed_dataset),
+          path(output_dir, stageAs: 'output/*'),
+          val(output_name)
+    val(ncpus)
+    val(mem_in_gb)
+
+    output:
+    tuple val(meta),
+          env(full_input_image_path),
+          val(input_dataset),
+          env(output_csv_file)      , emit: results
+    path "versions.yml"             , emit: versions
+
+    when:
+    task.ext.when == null || task.ext.when
+
+    script:
+    dapi_dataset_arg = dapi_dataset ? "--dapi-subpath ${dapi_dataset}" : ""
+    bleed_dataset_arg = bleed_dataset ? "--bleed-subpath ${bleed_dataset}" : ""
+    """
+    full_input_image_path=\$(readlink -e ${input_image_path})
+    full_labels_path=\$(readlink -e ${labels_path})
+    full_output_dir=\$(readlink ${output_dir})
+
+    mkdir -p \${full_output_dir}
+
+    output_csv_file="\${full_output_dir}/${output_name}"
+
+    CMD=(
+        python
+        /opt/scripts/spots-utils/labeled-spots-props.py
+        --image-container \${full_input_image_path}
+        --image-subpath ${input_dataset}
+        --labels-container \${full_labels_path}
+        --labels-subpath ${labels_dataset}
+        ${dapi_dataset_arg}
+        ${bleed_dataset_arg}
+        --output \${output_csv_file}
+    )
+    echo "CMD: \${CMD[@]}"
+    (exec "\${CMD[@]}")
+
+    cat <<-END_VERSIONS > versions.yml
+    "${task.process}":
+        measure-spots: v1
+    END_VERSIONS
+    """
+
+}

nextflow_schema.json

@@ -514,6 +514,166 @@
             },
             "help_text": "Registration parameters"
         },
+        "spot_extraction": {
+            "title": "Spot extraction",
+            "type": "object",
+            "description": "Spot extraction options",
+            "help_text": "Spot extraction arguments",
+            "fa_icon": "far fa-map",
+            "properties": {
+                "skip_spot_extraction": {
+                    "type": "boolean",
+                    "description": "If set skip spot extraction"
+                },
+                "spot_extraction_subdir": {
+                    "type": "string",
+                    "fa_icon": "fas fa-folder-open",
+                    "description": "Spot extraction results sub-directory"
+                },
+                "spot_extraction_ids": {
+                    "type": "string",
+                    "description": "rounds used for extracting spots"
+                },
+                "spot_channels": {
+                    "type": "string",
+                    "description": "Comma delimited list of spot channels. This together with spot_scales is used for determining the datasets used for spot extraction."
+                },
+                "spot_scales": {
+                    "type": "string",
+                    "description": "image scale used for spot extraction. This together with spot_channels is used for determining the datasets used for spot extraction."
+                },
+                "spot_subpaths": {
+                    "type": "string",
+                    "description": "alternative for defining the datasets used for spot extraction. If this is defined it takes priority over spot_channels and spot_scales."
+                },
+                "rsfish_min_intensity": {
+                    "type": "integer",
+                    "description": "Minimal intensity of the image. Default: 0",
+                    "fa_icon": "fas fa-angle-double-down"
+                },
+                "rsfish_max_intensity": {
+                    "type": "integer",
+                    "description": "Maximal intensity of the image. Default: 4096",
+                    "fa_icon": "fas fa-angle-double-up"
+                },
+                "rsfish_anisotropy": {
+                    "type": "number",
+                    "description": "The anisotropy factor. Default: 0.7",
+                    "help_text": "Scaling of z relative to xy. Can be determined using the RS-FISH anisotropy plugin in Fiji.",
+                    "fa_icon": "fas fa-arrows-alt"
+                },
+                "rsfish_sigma": {
+                    "type": "number",
+                    "description": "Sigma value for Difference-of-Gaussian (DoG) calculation. Default 1.5",
+                    "fa_icon": "fab fa-etsy"
+                },
+                "rsfish_threshold": {
+                    "type": "number",
+                    "description": "Threshold value for Difference-of-Gaussian (DoG) calculation. Default: 0.007",
+                    "fa_icon": "fas fa-level-down-alt"
+                },
+                "rsfish_background": {
+                    "type": "integer",
+                    "description": "Background subtraction method, 0 == None, 1 == Mean, 2==Median, 3==RANSAC on Mean, 4==RANSAC on Median. Default: 0 (None)",
+                    "fa_icon": "fas fa-level-down-alt"
+                },
+                "rsfish_intensity": {
+                    "type": "integer",
+                    "description": "Intensity calculation method, 0 == Linear Interpolation, 1 == Gaussian fit (on inlier pixels), 2 == Integrate spot intensities (on candidate pixels). Default: 0 (Linear Interpolation)",
+                    "fa_icon": "fas fa-level-down-alt"
+                },
+                "rsfish_spark_workers": {
+                    "type": "integer",
+                    "fa_icon": "fas fa-cogs",
+                    "description": "Number of Spark workers to use for spot extraction.",
+                    "default": 2
+                },
+                "rsfish_min_spark_workers": {
+                    "type": "integer",
+                    "fa_icon": "fas fa-cogs",
+                    "description": "Minimum number of Spark workers needed to start spot extraction pipeline.",
+                    "default": 1
+                },
+                "rsfish_spark_worker_cores": {
+                    "type": "integer",
+                    "fa_icon": "fas fa-microchip",
+                    "description": "Number of cores allocated to each Spark worker.",
+                    "default": 5
+                },
+                "rsfish_spark_gb_per_core": {
+                    "type": "integer",
+                    "fa_icon": "fas fa-cog",
+                    "description": "Size of memory (in GB) that is allocated for each core of a Spark worker.",
+                    "help_text": "The total memory usage for extracting spots from one acquisition will be workers * worker_cores * gb_per_core.",
+                    "default": 5
+                },
+                "rsfish_spark_driver_cores": {
+                    "type": "integer",
+                    "fa_icon": "fas fa-microchip",
+                    "description": "Number of cores allocated for the Spark driver. Default: 1",
+                    "default": 1
+                },
+                "rsfish_spark_driver_mem_gb": {
+                    "type": "integer",
+                    "fa_icon": "fas fa-memory",
+                    "description": "Amount of memory to allocate for the Spark driver. Default: 12g",
+                    "default": 4
+                }
+            }
+        },
+        "spots_features": {
+            "title": "Spots stats and characteristics",
+            "type": "object",
+            "description": "Various spots characteristcs, such as counts, area, intensity",
+            "fa_icon": "far fa-map",
+            "properties": {
+                "spots_counts_subdir": {
+                    "type": "string",
+                    "fa_icon": "fas fa-folder-open",
+                    "description": "data sub-directory used for spots counts"
+                },
+                "spots_props_subdir": {
+                    "type": "string",
+                    "fa_icon": "fas fa-folder-open",
+                    "description": "data sub-directory used for spots properties"
+                },
+                "dapi_channel": {
+                    "type": "string",
+                    "fa_icon": "fas fa-asterisk",
+                    "description": "Name of the DAPI channel.",
+                    "help_text": "The DAPI channel is used as a reference channel for registration, segmentation, and spot extraction."
+                },
+                "bleed_channel": {
+                    "type": "string",
+                    "fa_icon": "fas fa-asterisk",
+                    "description": "Channel (other than DAPI) that needs bleedthrough correction."
+                },
+                "spots_counts_cores": {
+                    "type": "integer",
+                    "fa_icon": "fas fa-microchip",
+                    "description": "Number of cores needed for getting spots counts.",
+                    "default": 1
+                },
+                "spots_counts_mem_gb": {
+                    "type": "integer",
+                    "fa_icon": "fas fa-memory",
+                    "description": "Amount of memory needed for getting spots counts",
+                    "default": 2
+                },
+                "spots_props_cores": {
+                    "type": "integer",
+                    "fa_icon": "fas fa-microchip",
+                    "description": "Number of cores needed for getting spots properties (intensity).",
+                    "default": 1
+                },
+                "spots_props_mem_gb": {
+                    "type": "integer",
+                    "fa_icon": "fas fa-memory",
+                    "description": "Amount of memory needed for getting spots properties (intensity)",
+                    "default": 2
+                }
+            }
+        },
         "institutional_config_options": {
             "title": "Institutional config options",
             "type": "object",
@@ -1030,6 +1190,12 @@
         {
             "$ref": "#/definitions/segmentation"
         },
+        {
+            "$ref": "#/definitions/spot_extraction"
+        },
+        {
+            "$ref": "#/definitions/spots_features"
+        },
         {
             "$ref": "#/definitions/generic_options"
         }