Initial check-in

README

@@ -0,0 +1,31 @@
+# bio.sam
+
+Clojure interface to SAM-JDK. Just at proof of concept stage, provides function
+to read SAM/BAM files and return lazy-seq of SAMRecord objects.
+
+## Usage
+
+(use '[bio.sam.core :only (read-sam)])
+
+(def s (read-sam "examples/toy.sam"))
+
+(map (memfn getReadString) s)
+
+## Installation
+
+FIXME: write
+
+## See also
+
+http://picard.sourceforge.net/
+
+## License
+
+Copyright (C) 2010, Ray Miller <ray@1729.org.uk>.
+
+THE ACCOMPANYING PROGRAM IS PROVIDED UNDER THE TERMS OF THIS ECLIPSE
+PUBLIC LICENSE, <http://www.eclipse.org/org/documents/epl-v10.php>.
+
+ANY USE, REPRODUCTION OR DISTRIBUTION OF THE PROGRAM CONSTITUTES
+RECIPIENT’S ACCEPTANCE OF THIS AGREEMENT.
+

examples/toy.sam

@@ -0,0 +1,7 @@
+@SQ	SN:ref	LN:45
+r001	163	ref	7	30	8M2I4M1D3M	=	37	39	TTAGATAAAGGATACTG	*
+r002	0	ref	9	30	1S2I6M1P1I4M2I	*	0	0	AAAAGATAAGGATAAA	*
+r003	0	ref	9	30	5H6M	*	0	0	AGCTAA	*
+r004	0	ref	16	30	6M14N1I5M	*	0	0	ATAGCTCTCAGC	*
+r003	16	ref	29	30	6H5M	*	0	0	TAGGC	*
+r001	83	ref	37	30	9M	=	7	-39	CAGCGCCAT	*

project.clj

@@ -0,0 +1,5 @@
+(defproject bio.sam "1.0.0-SNAPSHOT"
+  :description "FIXME: write"
+  :dependencies [[org.clojure/clojure "1.2.0-master-SNAPSHOT"]
+                 [org.clojure/clojure-contrib "1.2.0-SNAPSHOT"]]
+  :dev-dependencies [[swank-clojure "1.2.1"]])

src/bio/sam/core.clj

@@ -0,0 +1,43 @@
+(ns bio.sam.core
+  (:use [clojure.contrib.io :only (file)])
+  (:import [net.sf.samtools SAMFileReader SAMRecordIterator]))
+
+(defn read-sam [filename]
+  "Read file in SAM/BAM format and return a lazy sequence of
+   SAMRecord objects. Close the file when the last record has
+   been read"
+  (let [sam-reader (SAMFileReader. (file filename))
+	read-record (fn this [#^SAMRecordIterator it]
+		      (lazy-seq
+		       (if (.hasNext it)
+			 (cons (.next it) (this it))
+			 (do (.close it) (.close sam-reader)))))]
+    (read-record (.iterator sam-reader))))
+
+(comment
+
+  (def s (read-sam "examples/toy.sam"))
+  ;;=> #'user/s
+
+  (first s)
+  ;;=> #<SAMRecord r001 2/2 17b aligned read.>
+
+  (.getCigarString (first s))
+  ;;=> "8M2I4M1D3M"
+
+  (map (memfn getCigarString) s)
+  ;;=> ("8M2I4M1D3M" "1S2I6M1P1I4M2I" "5H6M" "6M14N1I5M" "6H5M" "9M")
+
+  (.getReadLength (first s))
+  ;;=> 17
+
+  (map (memfn getReadLength) s)
+  ;;=> (17 16 6 12 5 9)
+
+  (map (memfn getReadString) s)
+  ;;=> ("TTAGATAAAGGATACTG" "AAAAGATAAGGATAAA" "AGCTAA" "ATAGCTCTCAGC" "TAGGC" "CAGCGCCAT")
+
+  ;; See <http://picard.sourceforge.net/javadoc/net/sf/samtools/SAMRecord.html>
+  ;; for other methods supported by SAMRecord
+
+  )

test/bio/sam/core_test.clj

@@ -0,0 +1,6 @@
+(ns bio.sam.core-test
+  (:use [bio.sam.core] :reload-all)
+  (:use [clojure.test]))
+
+(deftest replace-me ;; FIXME: write
+  (is false))