Templates for SGN segmentation

flamingo_tools/data_conversion.py

@@ -10,7 +10,11 @@
 import numpy as np
 import pybdv
 
-from cluster_tools.utils.volume_utils import write_format_metadata
+try:
+    from cluster_tools.utils.volume_utils import write_format_metadata
+except ImportError:
+    write_format_metadata = None
+
 from elf.io import open_file
 from skimage.transform import rescale
 

reproducibility/templates_processing/REAMDE.md

@@ -0,0 +1,18 @@
+# Segmentation and detection workflows
+
+Implements workflows to segment SGNs or IHCs, and to detect ribbon synapses in slurm.
+
+For SGN segmentation run:
+- mean_std_SGN_template.sbatch
+- apply_unet_SGN_template.sbatch
+- segment_unet_SGN_template.sbatch
+
+For IHC segmentation run:
+- mean_std_IHC_template.sbatch
+- apply_unet_IHC_template.sbatch
+- segment_unet_IHC_template.sbatch
+
+For ribbon synapse detection without associated IHC segmentation run
+- detect_synapse_template.sbatch
+For ribbon synapse detection with associated IHC segmentation run
+- detect_synapse_marker_template.sbatch

reproducibility/templates_processing/apply_unet_IHC_template.sbatch

@@ -0,0 +1,60 @@
+#!/bin/bash
+#SBATCH --job-name=apply-unet-IHC
+#SBATCH -t 02:00:00             # for gerbil up to ~3 hours
+
+#SBATCH -p grete:shared         # the partition
+#SBATCH -G A100:1               # For requesting 1 A100 GPU.
+#SBATCH -c 1
+#SBATCH --mem 24G
+#SBATCH -a 0-9
+
+source ~/.bashrc
+# micromamba activate micro-sam_gpu
+micromamba activate sam
+
+# Print out some info.
+echo "Submitting job with sbatch from directory: ${SLURM_SUBMIT_DIR}"
+echo "Home directory: ${HOME}"
+echo "Working directory: $PWD"
+echo "Current node: ${SLURM_NODELIST}"
+
+# Run the script
+
+# SCRIPT_REPO=/user/schilling40/u15000/flamingo-tools
+SCRIPT_REPO=/user/pape41/u12086/Work/my_projects/flamingo-tools
+cd "$SCRIPT_REPO"/flamingo_tools/segmentation/ || exit
+
+export SCRIPT_DIR=$SCRIPT_REPO/scripts
+
+# name of cochlea, as it appears in MoBIE and the NHR
+COCHLEA=$1
+# data in n5 format, e.g. GEK11L_PV_GFP_01_fused.n5
+DATA=$2
+# channel in n5 folder, e.g. 0, 1, or 2
+STAIN_CHANNEL=$3
+# segmentation name, as it appears in MoBIE, e.g. IHC_v4
+SEG_NAME=$4
+
+export INPUT=/mnt/vast-nhr/projects/nim00007/data/moser/cochlea-lightsheet/"$COCHLEA"/"$DATA"
+
+export OUTPUT_FOLDER=/mnt/vast-nhr/projects/nim00007/data/moser/cochlea-lightsheet/predictions/"$COCHLEA"/"$SEG_NAME"
+
+# export MODEL=/mnt/vast-nhr/projects/nim00007/data/moser/cochlea-lightsheet/trained_models/IHC/v4_cochlea_distance_unet_IHC_supervised_2025-07-14
+export MODEL=/mnt/vast-nhr/projects/nim00007/data/moser/cochlea-lightsheet/trained_models/IHC/v5_cochlea_distance_unet_IHC_supervised_2025-08-20/
+export PREDICTION_INSTANCES=10
+export INPUT_KEY="setup$STAIN_CHANNEL/timepoint0/s0"
+
+echo "Input directory: ${INPUT}"
+echo "Output directory: ${OUTPUT_FOLDER}"
+echo "Model: ${MODEL}"
+
+cmd_array=(	'import sys,os;'
+	'sys.path.insert(0,os.environ["SCRIPT_DIR"]);'
+	'import unet_prediction;'
+	'unet_prediction.run_unet_prediction_slurm(input_path=os.environ["INPUT"],'
+	'output_folder=os.environ["OUTPUT_FOLDER"],model_path=os.environ["MODEL"],'
+	'input_key=os.environ["INPUT_KEY"],'
+	'prediction_instances=os.environ["PREDICTION_INSTANCES"])')
+cmd="${cmd_array[*]}"
+python -c "$cmd"
+

reproducibility/templates_processing/apply_unet_SGN_template.sbatch

@@ -0,0 +1,57 @@
+#!/bin/bash
+#SBATCH --job-name=apply-unet-SGN
+#SBATCH -t 02:00:00             # for gerbil up to ~3 hours
+
+#SBATCH -p grete:shared         # the partition
+#SBATCH -G A100:1               # For requesting 1 A100 GPU.
+#SBATCH -c 1
+#SBATCH --mem 24G
+#SBATCH -a 0-9
+
+source ~/.bashrc
+micromamba activate micro-sam_gpu
+
+# Print out some info.
+echo "Submitting job with sbatch from directory: ${SLURM_SUBMIT_DIR}"
+echo "Home directory: ${HOME}"
+echo "Working directory: $PWD"
+echo "Current node: ${SLURM_NODELIST}"
+
+# Run the script
+
+SCRIPT_REPO=/user/schilling40/u15000/flamingo-tools
+cd "$SCRIPT_REPO"/flamingo_tools/segmentation/ || exit
+
+export SCRIPT_DIR=$SCRIPT_REPO/scripts
+
+# name of cochlea, as it appears in MoBIE and the NHR
+COCHLEA=$1
+# data in n5 format, e.g. GEK11L_PV_GFP_01_fused.n5
+DATA=$2
+# channel in n5 folder, e.g. 0, 1, or 2
+STAIN_CHANNEL=$3
+# segmentation name, as it appears in MoBIE, e.g. SGN_v2 or Calb1_SGN_v2
+SEG_NAME=$4
+
+export INPUT=/mnt/vast-nhr/projects/nim00007/data/moser/cochlea-lightsheet/"$COCHLEA"/"$DATA"
+
+export OUTPUT_FOLDER=/mnt/vast-nhr/projects/nim00007/data/moser/cochlea-lightsheet/predictions/"$COCHLEA"/"$SEG_NAME"
+
+export MODEL=/mnt/vast-nhr/projects/nim00007/data/moser/cochlea-lightsheet/trained_models/SGN/v2_cochlea_distance_unet_SGN_supervised_2025-05-27
+export PREDICTION_INSTANCES=10
+export INPUT_KEY="setup$STAIN_CHANNEL/timepoint0/s0"
+
+echo "Input directory: ${INPUT}"
+echo "Output directory: ${OUTPUT_FOLDER}"
+echo "Model: ${MODEL}"
+
+cmd_array=(	'import sys,os;'
+	'sys.path.insert(0,os.environ["SCRIPT_DIR"]);'
+	'import unet_prediction;'
+	'unet_prediction.run_unet_prediction_slurm(input_path=os.environ["INPUT"],'
+	'output_folder=os.environ["OUTPUT_FOLDER"],model_path=os.environ["MODEL"],'
+	'input_key=os.environ["INPUT_KEY"],'
+	'prediction_instances=os.environ["PREDICTION_INSTANCES"])')
+cmd="${cmd_array[*]}"
+python -c "$cmd"
+

reproducibility/templates_processing/detect_synapse_marker_template.sbatch

@@ -0,0 +1,57 @@
+#!/bin/bash
+#SBATCH --job-name=synapse-marker
+#SBATCH -t 08:00:00             # estimated time, adapt to your needs
+#SBATCH [email protected] # change this to your mailaddress
+#SBATCH --mail-type=FAIL        # send mail when job begins and ends
+
+#SBATCH -p grete:shared         # the partition
+#SBATCH -G A100:1               # For requesting 1 A100 GPU.
+#SBATCH -A nim00007
+#SBATCH -c 2
+#SBATCH --mem 36G
+
+source ~/.bashrc
+micromamba activate micro-sam_gpu
+
+# Print out some info.
+echo "Submitting job with sbatch from directory: ${SLURM_SUBMIT_DIR}"
+echo "Home directory: ${HOME}"
+echo "Working directory: $PWD"
+echo "Current node: ${SLURM_NODELIST}"
+
+# Run the script
+#python myprogram.py $SLURM_ARRAY_TASK_ID
+
+SCRIPT_REPO=/user/schilling40/u15000/flamingo-tools
+cd "$SCRIPT_REPO"/flamingo_tools/segmentation/ || exit
+
+export SCRIPT_DIR=$SCRIPT_REPO/scripts
+
+# name of cochlea, as it appears in MoBIE and the NHR
+COCHLEA=$1
+# image channel, e.g. CTBP2 or RibA
+IMAGE_CHANNEL=$2
+# segmentation name, as it appears in MoBIE, e.g. synapses_v3
+IHC_SEG=$3
+
+export INPUT_PATH="$COCHLEA"/images/ome-zarr/"$IMAGE_CHANNEL".ome.zarr
+export MASK_PATH="$COCHLEA"/images/ome-zarr/"$IHC_SEG".ome.zarr
+
+export OUTPUT_FOLDER=/mnt/vast-nhr/projects/nim00007/data/moser/cochlea-lightsheet/predictions/"$COCHLEA"/synapses_v3_"$IHC_SEG"
+
+export MODEL=/mnt/vast-nhr/projects/nim00007/data/moser/cochlea-lightsheet/trained_models/Synapses/synapse_detection_model_v3.pt
+export INPUT_KEY="s0"
+export MAX_DISTANCE=8
+
+echo "OUTPUT_FOLDER $OUTPUT_FOLDER"
+echo "MODEL $MODEL"
+
+python ~/flamingo-tools/scripts/synapse_marker_detection/marker_detection.py \
+	--input "$INPUT_PATH" \
+	--input_key $INPUT_KEY \
+	--output_folder "$OUTPUT_FOLDER" \
+	--mask "$MASK_PATH" \
+	--model $MODEL \
+	--max_distance $MAX_DISTANCE \
+	--s3
+

reproducibility/templates_processing/detect_synapse_template.sbatch

@@ -0,0 +1,61 @@
+#!/bin/bash
+#SBATCH --job-name=synapse-detect
+#SBATCH -t 42:00:00             # estimated time, adapt to your needs
+#SBATCH [email protected] # change this to your mailaddress
+#SBATCH --mail-type=FAIL        # send mail when job begins and ends
+
+#SBATCH -p grete:shared         # the partition
+#SBATCH -G A100:1               # For requesting 1 A100 GPU.
+#SBATCH -A nim00007
+#SBATCH -c 2
+#SBATCH --mem 500G
+
+source ~/.bashrc
+micromamba activate micro-sam_gpu
+
+# Print out some info.
+echo "Submitting job with sbatch from directory: ${SLURM_SUBMIT_DIR}"
+echo "Home directory: ${HOME}"
+echo "Working directory: $PWD"
+echo "Current node: ${SLURM_NODELIST}"
+
+# Run the script
+#python myprogram.py $SLURM_ARRAY_TASK_ID
+
+SCRIPT_REPO=/user/schilling40/u15000/flamingo-tools
+cd "$SCRIPT_REPO"/flamingo_tools/segmentation/ || exit
+
+export SCRIPT_DIR=$SCRIPT_REPO/scripts
+
+# name of cochlea, as it appears in MoBIE and the NHR
+COCHLEA=$1
+# image channel, e.g. CTBP2 or RibA
+IMAGE_CHANNEL=$2
+# segmentation name, as it appears in MoBIE, e.g. synapses_v3
+IHC_SEG=$3
+
+export INPUT_PATH="$COCHLEA"/images/ome-zarr/"$IMAGE_CHANNEL".ome.zarr
+export MASK_PATH="$COCHLEA"/images/ome-zarr/"$IHC_SEG".ome.zarr
+
+# data on NHR
+# export INPUT_PATH=/mnt/vast-nhr/projects/nim00007/data/moser/cochlea-lightsheet/"$COCHLEA"/"$DATA"
+# export INPUT_KEY="setup$STAIN_CHANNEL/timepoint0/s0"
+
+export OUTPUT_FOLDER=/mnt/vast-nhr/projects/nim00007/data/moser/cochlea-lightsheet/predictions/"$COCHLEA"/synapses_v3
+
+if ! [[ -f $OUTPUT_FOLDER ]] ; then
+	mkdir -p "$OUTPUT_FOLDER"
+fi
+
+export MODEL=/mnt/vast-nhr/projects/nim00007/data/moser/cochlea-lightsheet/trained_models/Synapses/synapse_detection_model_v3.pt
+
+echo "OUTPUT_FOLDER $OUTPUT_FOLDER"
+echo "MODEL $MODEL"
+
+python ~/flamingo-tools/scripts/synapse_marker_detection/run_prediction.py \
+	--input "$INPUT_PATH" \
+	--input_key "$INPUT_KEY" \
+	--output_folder "$OUTPUT_FOLDER" \
+	--model $MODEL \
+	--s3
+

reproducibility/templates_processing/mean_std_IHC_template.sbatch

@@ -0,0 +1,51 @@
+#!/bin/bash
+#SBATCH --job-name=mean-std-IHC
+#SBATCH -t 01:00:00             # usually takes ~20 min
+
+#SBATCH -p standard96s:shared         # the partition
+#SBATCH -A nim00007
+#SBATCH -c 3
+#SBATCH --mem 128G
+
+source ~/.bashrc
+# micromamba activate flamingo13
+micromamba activate sam
+
+# Run the script
+
+# SCRIPT_REPO=/user/schilling40/u15000/flamingo-tools
+SCRIPT_REPO=/user/pape41/u12086/Work/my_projects/flamingo-tools
+cd "$SCRIPT_REPO"/flamingo_tools/segmentation/ || exit
+
+export SCRIPT_DIR=$SCRIPT_REPO/scripts
+
+# name of cochlea, as it appears in MoBIE and the NHR
+COCHLEA=$1
+# data in n5 format, e.g. GEK11L_PV_GFP_01_fused.n5
+DATA=$2
+# channel in n5 folder, e.g. 0, 1, or 2
+STAIN_CHANNEL=$3
+# segmentation name, as it appears in MoBIE, e.g. IHC_v4
+SEG_NAME=$4
+
+export INPUT=/mnt/vast-nhr/projects/nim00007/data/moser/cochlea-lightsheet/"$COCHLEA"/"$DATA"
+export OUTPUT_FOLDER=/mnt/vast-nhr/projects/nim00007/data/moser/cochlea-lightsheet/predictions/"$COCHLEA"/"$SEG_NAME"
+export SEG_CLASS="ihc"
+export INPUT_KEY="setup$STAIN_CHANNEL/timepoint0/s0"
+
+if ! [[ -f $OUTPUT_FOLDER ]] ; then
+	mkdir -p "$OUTPUT_FOLDER"
+fi
+
+echo "Input directory: ${INPUT}"
+echo "Output directory: ${OUTPUT_FOLDER}"
+
+cmd_array=(	'import sys,os;'
+	'sys.path.insert(0,os.environ["SCRIPT_DIR"]);'
+	'import unet_prediction;'
+	'unet_prediction.run_unet_prediction_preprocess_slurm(input_path=os.environ["INPUT"],'
+	'input_key=os.environ["INPUT_KEY"],'
+	'output_folder=os.environ["OUTPUT_FOLDER"],seg_class=os.environ["SEG_CLASS"])')
+cmd="${cmd_array[*]}"
+python -c "$cmd"
+

reproducibility/templates_processing/mean_std_SGN_template.sbatch

@@ -0,0 +1,49 @@
+#!/bin/bash
+#SBATCH --job-name=mean-std-SGN
+#SBATCH -t 01:00:00             # usually takes ~20 min
+
+#SBATCH -p standard96s:shared         # the partition
+#SBATCH -A nim00007
+#SBATCH -c 3
+#SBATCH --mem 128G
+
+source ~/.bashrc
+micromamba activate flamingo13
+
+# Run the script
+
+SCRIPT_REPO=/user/schilling40/u15000/flamingo-tools
+cd "$SCRIPT_REPO"/flamingo_tools/segmentation/ || exit
+
+export SCRIPT_DIR=$SCRIPT_REPO/scripts
+
+# name of cochlea, as it appears in MoBIE and the NHR
+COCHLEA=$1
+# data in n5 format, e.g. GEK11L_PV_GFP_01_fused.n5
+DATA=$2
+# channel in n5 folder, e.g. 0, 1, or 2
+STAIN_CHANNEL=$3
+# segmentation name, as it appears in MoBIE, e.g. SGN_v2 or Calb1_SGN_v2
+SEG_NAME=$4
+
+export INPUT=/mnt/vast-nhr/projects/nim00007/data/moser/cochlea-lightsheet/"$COCHLEA"/"$DATA"
+export OUTPUT_FOLDER=/mnt/vast-nhr/projects/nim00007/data/moser/cochlea-lightsheet/predictions/"$COCHLEA"/"$SEG_NAME"
+export SEG_CLASS="sgn"
+export INPUT_KEY="setup$STAIN_CHANNEL/timepoint0/s0"
+
+if ! [[ -f $OUTPUT_FOLDER ]] ; then
+	mkdir -p "$OUTPUT_FOLDER"
+fi
+
+echo "Input directory: ${INPUT}"
+echo "Output directory: ${OUTPUT_FOLDER}"
+
+cmd_array=(	'import sys,os;'
+	'sys.path.insert(0,os.environ["SCRIPT_DIR"]);'
+	'import unet_prediction;'
+	'unet_prediction.run_unet_prediction_preprocess_slurm(input_path=os.environ["INPUT"],'
+	'input_key=os.environ["INPUT_KEY"],'
+	'output_folder=os.environ["OUTPUT_FOLDER"],seg_class=os.environ["SEG_CLASS"])')
+cmd="${cmd_array[*]}"
+python -c "$cmd"
+