1 // ***************************************************************************
2 // BamAlignment.h (c) 2009 Derek Barnett
3 // Marth Lab, Department of Biology, Boston College
4 // All rights reserved.
5 // ---------------------------------------------------------------------------
6 // Last modified: 13 December 2010 (DB)
7 // ---------------------------------------------------------------------------
8 // Provides the BamAlignment data structure
9 // ***************************************************************************
11 #ifndef BAMALIGNMENT_H
12 #define BAMALIGNMENT_H
14 #include <api/api_global.h>
15 #include <api/BamAux.h>
21 // forward declare BamAlignment's friend classes
23 class BamReaderPrivate;
24 class BamWriterPrivate;
25 } // namespace Internal
27 // BamAlignment data structure
28 // explicitly labeled as 'struct' to indicate that (most of) its fields are public
29 struct API_EXPORT BamAlignment {
31 // constructors & destructor
34 BamAlignment(const BamAlignment& other);
37 // Queries against alignment flags
39 bool IsDuplicate(void) const; // Returns true if this read is a PCR duplicate
40 bool IsFailedQC(void) const; // Returns true if this read failed quality control
41 bool IsFirstMate(void) const; // Returns true if alignment is first mate on read
42 bool IsMapped(void) const; // Returns true if alignment is mapped
43 bool IsMateMapped(void) const; // Returns true if alignment's mate is mapped
44 bool IsMateReverseStrand(void) const; // Returns true if alignment's mate mapped to reverse strand
45 bool IsPaired(void) const; // Returns true if alignment part of paired-end read
46 bool IsPrimaryAlignment(void) const; // Returns true if reported position is primary alignment
47 bool IsProperPair(void) const; // Returns true if alignment is part of read that satisfied paired-end resolution
48 bool IsReverseStrand(void) const; // Returns true if alignment mapped to reverse strand
49 bool IsSecondMate(void) const; // Returns true if alignment is second mate on read
51 // Manipulate alignment flags
53 void SetIsDuplicate(bool ok); // Sets "PCR duplicate" flag
54 void SetIsFailedQC(bool ok); // Sets "failed quality control" flag
55 void SetIsFirstMate(bool ok); // Sets "alignment is first mate" flag
56 void SetIsMapped(bool ok); // Sets "alignment is mapped" flag
57 void SetIsMateMapped(bool ok); // Sets "alignment's mate is mapped" flag
58 void SetIsMateReverseStrand(bool ok); // Sets "alignment's mate mapped to reverse strand" flag
59 void SetIsPaired(bool ok); // Sets "alignment part of paired-end read" flag
60 void SetIsPrimaryAlignment(bool ok); // Sets "position is primary alignment" flag
61 void SetIsProperPair(bool ok); // Sets "alignment is part of read that satisfied paired-end resolution" flag
62 void SetIsReverseStrand(bool ok); // Sets "alignment mapped to reverse strand" flag
63 void SetIsSecondMate(bool ok); // Sets "alignment is second mate on read" flag
65 // legacy methods (deprecated, but available)
66 void SetIsMateUnmapped(bool ok); // Complement of IsMateMapped() flag
67 void SetIsSecondaryAlignment(bool ok); // Complement of IsPrimaryAlignment() flag
68 void SetIsUnmapped(bool ok); // Complement of IsMapped() flag
70 // Tag data access methods
72 // -------------------------------------------------------------------------------------
73 // N.B. - The following tag access methods may not be used on BamAlignments fetched
74 // using BamReader::GetNextAlignmentCore(). Attempting to use them will not result in
75 // error message (to keep output clean) but will ALWAYS return false. Only user-created
76 // BamAlignments or those retrieved using BamReader::GetNextAlignment() are valid here.
78 // add tag data (create new TAG entry with TYPE and VALUE)
79 // TYPE is one of {A, i, f, Z, H} depending on VALUE - see SAM/BAM spec for details
80 // returns true if new data added, false if error or TAG already exists
81 // N.B. - will NOT modify existing tag. Use EditTag() instead
82 // @tag - two character tag name
83 // @type - single character tag type (see SAM/BAM spec for details)
84 // @value - value to associate with tag
85 bool AddTag(const std::string& tag, const std::string& type, const std::string& value); // type must be Z or H
86 bool AddTag(const std::string& tag, const std::string& type, const uint32_t& value); // type must be A or i
87 bool AddTag(const std::string& tag, const std::string& type, const int32_t& value); // type must be A or i
88 bool AddTag(const std::string& tag, const std::string& type, const float& value); // type must be A, i, or f
90 // edit tag data (sets existing TAG with TYPE to VALUE or adds new TAG if not already present)
91 // TYPE is one of {A, i, f, Z, H} depending on VALUE - see SAM/BAM spec for details
92 // returns true if edit was successfaul, false if error
93 // @tag - two character tag name
94 // @type - single character tag type (see SAM/BAM spec for details)
95 // @value - new value for tag
96 bool EditTag(const std::string& tag, const std::string& type, const std::string& value); // type must be Z or H
97 bool EditTag(const std::string& tag, const std::string& type, const uint32_t& value); // type must be A or i
98 bool EditTag(const std::string& tag, const std::string& type, const int32_t& value); // type must be A or i
99 bool EditTag(const std::string& tag, const std::string& type, const float& value); // type must be A, i, or f
101 // specific tag data access methods - these only remain for legacy support
102 // returns whether specific tag could be retrieved
103 bool GetEditDistance(uint32_t& editDistance) const; // get "NM" tag data (equivalent to GetTag("NM", editDistance))
104 bool GetReadGroup(std::string& readGroup) const; // get "RG" tag data (equivalent to GetTag("RG", readGroup))
106 // generic tag data access methods
107 // returns whether tag is found & tag type is compatible with DESTINATION
108 // @tag - two character tag name
109 // @destination - if found, tag value is stored here
110 bool GetTag(const std::string& tag, std::string& destination) const; // access variable-length char or hex strings
111 bool GetTag(const std::string& tag, uint32_t& destination) const; // access unsigned integer data
112 bool GetTag(const std::string& tag, int32_t& destination) const; // access signed integer data
113 bool GetTag(const std::string& tag, float& destination) const; // access floating point data
115 // retrieve the tag type code for TAG
116 // returns true if tag could be found and type determined
117 bool GetTagType(const std::string& tag, char& type) const;
120 // returns true if removal was successful, false if error
121 // N.B. - returns false if TAG does not exist (no removal can occur)
122 // @tag - two character tag name
123 bool RemoveTag(const std::string& tag);
125 // Additional data access methods
127 // calculates & returns alignment end position, based on starting position and CIGAR operations
128 // @usePadded - if true, counts inserted bases. Default is false, so that alignment end position matches the last base's position in reference
129 // @zeroBased - if true, returns 0-based coordinate; else returns 1-based. Setting this to false is useful when using BAM data along with other, half-open formats.
130 int GetEndPosition(bool usePadded = false, bool zeroBased = true) const;
132 // 'internal' utility methods
134 static bool FindTag(const std::string& tag, char* &pTagData, const unsigned int& tagDataLength, unsigned int& numBytesParsed);
135 static bool SkipToNextTag(const char storageType, char* &pTagData, unsigned int& numBytesParsed);
139 std::string Name; // Read name
140 int32_t Length; // Query length
141 std::string QueryBases; // 'Original' sequence (as reported from sequencing machine)
142 std::string AlignedBases; // 'Aligned' sequence (includes any indels, padding, clipping)
143 std::string Qualities; // FASTQ qualities (ASCII characters, not numeric values)
144 std::string TagData; // Tag data (accessor methods will pull the requested information out)
145 int32_t RefID; // ID number for reference sequence
146 int32_t Position; // Position (0-based) where alignment starts
147 uint16_t Bin; // Bin in BAM file where this alignment resides
148 uint16_t MapQuality; // Mapping quality score
149 uint32_t AlignmentFlag; // Alignment bit-flag - see Is<something>() methods to query this value, SetIs<something>() methods to manipulate
150 std::vector<CigarOp> CigarData; // CIGAR operations for this alignment
151 int32_t MateRefID; // ID number for reference sequence where alignment's mate was aligned
152 int32_t MatePosition; // Position (0-based) where alignment's mate starts
153 int32_t InsertSize; // Mate-pair insert size
155 // Internal data, inaccessible to client code
156 // but available BamReaderPrivate & BamWriterPrivate
158 struct BamAlignmentSupportData {
161 std::string AllCharData;
162 uint32_t BlockLength;
163 uint32_t NumCigarOperations;
164 uint32_t QueryNameLength;
165 uint32_t QuerySequenceLength;
169 BamAlignmentSupportData(void)
171 , NumCigarOperations(0)
173 , QuerySequenceLength(0)
177 BamAlignmentSupportData SupportData;
178 friend class Internal::BamReaderPrivate;
179 friend class Internal::BamWriterPrivate;
181 // Alignment flag query constants
182 // Use the get/set methods above instead
198 // convenience typedef(s)
199 typedef std::vector<BamAlignment> BamAlignmentVector;
201 } // namespace BamTools
203 #endif // BAMALIGNMENT_H