A Makefile to run the taxonomic classification tool for sequences, Kraken, to filter and to create reports of the results.

Kraken.mk

SHELL := /bin/bash

#####
# PRESETS
#####

FASTA?=YOUR-FASTA-FILE.fa
KRAKEN_PATH?=KRAKEN-BINARY-TOP-FOLDER
KRAKEN_CORES?=10
KRAKEN_FILTER_THRESHOLD?=0.2

KRAKEN_BIN=$(KRAKEN_PATH)/kraken
KRAKEN_FILTER_BIN=$(KRAKEN_PATH)/kraken-filter
KRAKEN_REPORT_BIN=$(KRAKEN_PATH)/kraken-report
KRAKEN_DB?=PATH-TO-KRAKEN-DB

#####
# BEAUTY TARGETS
#####
.PHONY: all kraken kraken_filter kraken_report kraken_filter_report

all: kraken kraken_filter kraken_report kraken_filter_report

kraken: $(FASTA:.fa=.kraken)

kraken_filter: $(FASTA:.fa=.kraken_filter)

kraken_report: $(FASTA:.fa=.kraken_report)

kraken_filter_report: $(FASTA:.fa=.kraken_filter_report)

#####
# ACTUAL TARGETS
#####

# Run kraken
%.kraken: %.fa
        @echo "Running Kraken ..."
        @date
        $(KRAKEN_BIN) --db $(KRAKEN_DB) --threads $(KRAKEN_CORES) $^ > $@
        @echo "done (kraken)."
        @date

# Run filter
%.kraken_filter: %.kraken
        @echo "Running Kraken-Filter $(KRAKEN_THRESHOLD) ..."
        @date
        $(KRAKEN_FILTER_BIN) --db $(KRAKEN_DB) --threshold $(KRAKEN_FILTER_THRESHOLD) $^ > $@
        @echo "done (kraken-filter)."
        @date

# Create a report
%_report: %
        @echo "Running Kraken-Report ..."
        @date
        $(KRAKEN_REPORT_BIN) --db $(KRAKEN_DB) $^ > $@
        @echo "done (kraken-report)."
        @date
#####
# CLEAN-UP
#####
clean:
        rm -f *.kraken*

Kraken_incl_db_name_appended.mk

SHELL := /bin/bash

#####
# PRESETS
#####

FASTA?=YOUR-FASTA-FILE.fa
KRAKEN_PATH?=KRAKEN-BINARY-TOP-FOLDER
KRAKEN_CORES?=10
KRAKEN_FILTER_THRESHOLD?=0.2

KRAKEN_BIN=$(KRAKEN_PATH)/kraken
KRAKEN_FILTER_BIN=$(KRAKEN_PATH)/kraken-filter
KRAKEN_REPORT_BIN=$(KRAKEN_PATH)/kraken-report
KRAKEN_DB?=PATH-TO-KRAKEN-DB
DB=$(notdir $(KRAKEN_DB))

#####
# BEAUTY TARGETS
#####
.PHONY: all kraken kraken_filter kraken_report kraken_filter_report

all: kraken kraken_filter kraken_report kraken_filter_report

kraken: $(FASTA:.fa=_x_$(DB).kraken)

kraken_filter: $(FASTA:.fa=_x_$(DB).kraken_filter)

kraken_report: $(FASTA:.fa=_x_$(DB).kraken_report)

kraken_filter_report: $(FASTA:.fa=_x_$(DB).kraken_filter_report)

#####
# ACTUAL TARGETS
#####

# Run kraken
%_x_$(DB).kraken: %.fa
        @echo "Running Kraken against microbes DB ..."
        @date
        $(KRAKEN_BIN) --db $(KRAKEN_DB) --threads $(KRAKEN_CORES) $^ > $@
        @echo "done (kraken against microbes db)."
        @date

# Run filter
%.kraken_filter: %.kraken
        @echo "Running Kraken-Filter $(KRAKEN_THRESHOLD) ..."
        @date
        $(KRAKEN_FILTER_BIN) --db $(KRAKEN_DB) --threshold $(KRAKEN_FILTER_THRESHOLD) $^ > $@
        @echo "done (kraken-filter)."
        @date

# Create report
%_report: %
        @echo "Running Kraken-Report ..."
        @date
        $(KRAKEN_REPORT_BIN) --db $(KRAKEN_DB) $^ > $@
        @echo "done (kraken-report)."
        @date

#####
# CLEAN-UP
#####
clean:
        rm -f *.kraken*

N.B. Make sure that the indentation of the recipes commands are all tab-indented.

To enable running the Makefile on your installation, you need to adjust/set the variables:

• FASTA: The path to your fasta-file of interest (needs to end in .fa, here)
• KRAKEN_PATH: The path your local Kraken-installation, i.e., the top directory including executables kraken, kraken-filter, and kraken-report.
• KRAKEN_DB: The path to your Kraken database.

You can either adjust these variable within the Makefile, or your could do it within the make call, e.g., make -f Kraken.mk FASTA=PATH-TO-YOUR-FASTA-FILE KRAKEN_PATH=YOUR-KRAKEN-PATH KRAKEN_DB=YOUR-KRAKEN-DB-PATH.

Kraken_incl_db_name_appended.mk serves the same purpose as Kraken.mk but additionally includes the name of the database in the output. This is particularly helpful if the same set of sequences should be searched against different databases, e.g., databases that were constructed using different k-mer sizes.