def setup_ripple(num_top):
config = {
"region": "us-west-2",
"role": "service-role/lambdaFullAccessRole",
"memory_size": 3008
}
pipeline = ripple.Pipeline(name="tide",
table="s3://maccoss-tide",
log="s3://maccoss-log",
timeout=600,
config=config)
params = {"database_bucket": "maccoss-fasta"}
input = pipeline.input(format="mzML")
# Filter for the most intense spectra
step = input.top(identifier="tic", number=num_top)
# Run param-medic on top spectra
step = input.run("parammedic", params=params, output_format="mzML")
# For each matching FASTA, run Tide
tide = lambda input_key, bucket_key: input_key.run(
"tide",
params={
"database_bucket": "maccoss-fasta",
"num_threads": 0,
"output_format": "tsv",
"species": bucket_key,
})
step = step.map(func=tide, params={"directories": True})
# Calculate the confidence score of the results
step = step.run("confidence", params=params, output_format="confidence")
# Find the result with the top confidence score
step = step.match("qvalue")
pipeline.compile("json/medic-tide.json", dry_run=False)
def setup(table, log, json_name, split_size, mem_size, region):
config = {
"region": region,
"role": "service-role/lambdaFullAccessRole",
"memory_size": 3008
}
pipeline = ripple.Pipeline(name="tide",
table="s3://" + table,
log="s3://" + log,
timeout=600,
config=config)
input = pipeline.input(format="mzML")
step = input.split({"split_size": split_size}, {"memory_size": 128})
params = {
"database_bucket": "maccoss-fasta",
"num_threads": 0,
"species": "normalHuman",
}
step = input.run("tide",
params=params,
output_format="tsv",
config={"memory_size": mem_size})
step = step.combine(params={"sort": False}, config={"memory_size": 256})
params = {
"database_bucket": "maccoss-fasta",
"max_train": 10 * 1000,
"output": "peptides",
}
step = step.run("percolator", params=params)
pipeline.compile(json_name, dry_run=False)
return len(pipeline.pipeline)
import ripple
config = {
"region": "us-west-2",
"role": "serverless-role",
"memory_size": 3008
}
pipeline = ripple.Pipeline(name="spacenet", table="s3://spacenet", log="s3://log", timeout=600, config=config)
input = pipeline.input(format="tif")
step = input.run("convert_to_pixels", params={"pixels_per_bin": 1000}, output_format="pixel")
step = step.run("pair", params={"split_size": 10*1000*1000})
step = step.run("run_knn", {"k": 100}, output_format="knn")
step = step.combine(params={"k": 100, "sort": True})
step = step.combine(params={"k": 100, "sort": False})
step = step.run("draw_borders", {"image": input}, output_format="tif")
pipeline.compile("json/spacenet-classification.json")
import ripple
config = {
"region": "us-west-2",
"role": "service-role/lambdaFullAccessRole",
"memory_size": 1024,
}
pipeline = ripple.Pipeline(name="tide",
table="s3://maccoss-tide",
log="s3://maccoss-log",
timeout=600,
config=config)
input = pipeline.input(format="new_line")
step = input.run("echo")
pipeline.compile("json/simple.json", dry_run=False)
import ripple
config = {
"region": "us-east-1",
"role": "serverless-role",
"memory_size": 3008,
}
pipeline = ripple.Pipeline(name="compression", table="s3://compression", log="s3://log", timeout=600, config=config)
input = pipeline.input(format="bed")
step = input.sort(identifier="start_position", params={"split_size": 500*1000*1000}, config={"memory_size": 3008})
step = step.run("compress_methyl", params={"program_bucket": "program"})
pipeline.compile("json/compression.json")
import ripple
config = {
"region": "us-west-2",
"role": "service-role/lambdaFullAccessRole",
"memory_size": 3008,
}
pipeline = ripple.Pipeline(name="compression", table="s3://maccoss-tide-west-2", log="s3://maccoss-log-west-2", timeout=600, config=config)
input = pipeline.input(format="bed")
step = input.sort(identifier="start_position", params={"split_size": 500*1000*1000, "num_bins": 35}, config={"memory_size": 3008})
step = step.run("compress_methyl", params={"program_bucket": "maccoss-methyl-data"})
pipeline.compile("json/compression.json")
import ripple
config = {
"region": "us-east-1",
"role": "service-role/lambdaFullAccessRole",
"memory_size": 3008,
}
pipeline = ripple.Pipeline(name="document",
table="s3://maccoss-tide-east-1",
log="s3://maccoss-log-east-1",
timeout=600,
config=config)
input = pipeline.input(format="new_line")
input.map(table="train-data",
func=lambda input_key, bucket_key: input_key.run(
"compress_fastore", params={"train_data": bucket_key}))
pipeline.compile("json/document.json", dry_run=True)
