Compressing BAM or CRAM files¶
Compressing a BAM or SAM file
$ genozip myfile.bam
genozip myfile.bam : Done (2 seconds, BAM compression ratio: 2.2)
$ ls -lh myfile.bam*
-rwxrwxrwx 1 divon divon 5.7M Aug 20 13:59 myfile.bam
-rwxrwxrwx 1 divon divon 2.6M Aug 20 13:59 myfile.bam.genozip
This creates a compressed file, without modifying the original file. This also works with .sam, .sam.gz, .sam.bz2 and .sam.xz files.
Some useful command line options (for a full list, see genozip manual):
genozip --test myfile.bam: after completing the compression, the file is uncompressed in memory, and its MD5 is compared to that of the original file.
genozip --replace myfile.bam: the original file is removed after successful compression
genozip --optimize-QUAL myfile.bam: the QUAL field (base quality scores) is optimized to improve compression. See genozip manual for details.
Compressing multiple files into a tar archive
genozip *.bam --tar mydata.tar. See details: Archiving.
Uncompressing
genounzip myfile.bam.genozip
Uncompresses a file.
genocat myfile.bam.genozip
Uncompresses a file into stdout (i.e. the terminal).
genounzip --index myfile.bam.genozip
Uncompresses a file and also generates a BAI index file, using samtools index. samtools needs to be installed for this option to work.
genounzip --output newname.sam.gz myfile.sam.genozip
genocat --output newname.bam myfile.bam.genozip
Uncompresses to a particular name. The file format produced depends on whether the output file name extension is .bam, .sam or .sam.gz. Converting between SAM and BAM is possible with genocat but not genounzip.
genocat --sam myfile.bam.genozip
genocat --bam myfile.sam.genozip
Specifies a SAM or BAM file output (implicit if --output is used).
genocat --bgzf 6 myfile.bam.genozip
genounzip --bgzf 6 myfile.bam.genozip
Sets the level BGZF compression (for .bam and .sam.gz output format) - from 0 (no compression) to 12 (best yet slowest compression). Absent this option, genounzip attemps to recover the BGZF compression level of the original file, while genocat uncompresses without BGZF compression for SAM and with BGZF compression level 1 for BAM (this is because some popular bioinforamtics tools error on BAM files with BGZF compression level 0).
Using in a pipeline
--input sam for a SAM file)my-pipeline | genozip - --input bam --output myfile.bam.genozipgenocat myfile.bam.genozip | my-pipelineUsing a reference file
Compressing against a reference file improves compression, often considerably, and is therefore recommended:
$ genozip --make-reference hs37d5.fa.gz # one-time preparation of the reference. should be similar to the reference used to create the BAM file
$ genozip myfile.bam --reference hs37d5.ref.genozip
Note: the reference file is also needed for uncompressing. Alternatively, use --REFERENCE (uppercase) which copies the relevant parts of the reference file to myfile.bam, obliviating the need for a reference file for uncompressing. Compression with a reference file works regardless of whether the SAM/BAM file is aligned.
Compressing CRAM files
Genozip does not support CRAM natively - it uses samtools to convert CRAM files to the BAM format, and as such it requires samtools to be installed for CRAM compression to work.
genozip --reference hs37d5.ref.genozip myfile.cram
Compresses a CRAM file into myfile.sam.genozip in this case.
Note: use of a reference is mandatory when compressing a CRAM file.
Note: When decompressing the file, it is decompressed into SAM or BAM format. Genozip does not support decompressing to CRAM format.
Compressing Ion Torrent BAM files
Genozip offers the --optimize-ZM option to optimize the ZM (flow signal) data for improved compression. See genozip manual for details.
Example:
$ wget ftp://ftp-trace.ncbi.nih.gov/HG002_NA24385_SRR1767406_IonXpress_020_rawlib_24028.30k.b37.bam
$ genozip IonXpress_020_rawlib.b37.bam
$ genozip IonXpress_020_rawlib.b37.bam --optimize-ZM -o IonXpress_020_rawlib.b37.optimized.bam.genozip
$ ls -Ggh IonXpress_020_rawlib.b37*
-rw-rw-r--+ 1 26G Aug 13 23:53 IonXpress_020_rawlib.b37.bam
-rw-rw-r--+ 1 17G Aug 14 00:10 IonXpress_020_rawlib.b37.bam.genozip
-rw-rw-r--+ 1 12G Aug 14 00:17 IonXpress_020_rawlib.b37.optimized.bam.genozip
Best compression
Using the --best option causes Genozip to use more aggressive compression methods, at the expense of higher CPU and memory usage, resulting in better compression.
genozip --best myfile.bam --reference hs37d5.ref.genozip
Fast compression
Using the --fast option causes Genozip compress faster, at the expense of a lower compression ratio. This option also usually results in faster decompression and lower memory consumption.
genozip --fast myfile.bam
Converting to a FASTQ
genocat --fastq myfile.bam.genozip or genozip --fastq=all myfile.bam.genozip may be used to output the data in FASTQ format. See Converting SAM/BAM to FASTQ for details.
Downsampling
genocat --downsample 10,0 myfile.bam.genozip
Displays only the first (#0) read in every 10 reads.
Grepping
genocat --grep-w MC:Z:151M myfile.bam.genozip
Displays the lines containing “MC:Z:151M” (strings that match exactly).
genocat --grep ACCTTAAT myfile.bam.genozip
Displays the lines containing “ACCTTAAT” (possibly a substring of a longer string).
The SAM header
genocat --header-only myfile.bam.genozip
Displays only the SAM header.
genocat --no-header myfile.bam.genozip
Displays the file without the SAM header.
genocat --no-PG myfile.bam.genozip
When modifying the data in a file using genocat, Genozip normally adds a @PG line to the header with information about the modification. With this option it doesn’t.
Filtering specific regions of the genome
Examples of using --regions (or its shortcut -r):
|
Positions 1000 to 2000 on contig 22 |
|
151 bases, starting pos 1000, on contig 22 |
|
Two ranges on all contigs |
|
Contigs chr21 and chr22 in their entirety |
|
All contigs, excluding MT and Y |
|
All contigs, excluding positions up to 1000 |
|
Contig chrM |
genocat --regions-file <filename>
Get regions from a tab-separated file. An example of a valid file:
chr22 17000000 17000099
chr22 17000000 +100
chr22 17000000
Filtering reads based on FLAG
genocat --FLAG *{+-^}value* myfile.bam.genozip. Filter lines based on the FLAG value: <value> is a decimal or hexadecimal value and should be prefixed by + - or ^:
+
INCLUDES lines in which ALL flags in value are set in the line’s FLAG
-
INCLUDES lines in which NO flags in value are set in the line’s FLAG
^
EXCLUDES lines in which ALL flags in value are set in the line’s FLAG
Example: genocat --FLAG -192 includes only lines in which neither FLAG 64 nor 128 are set. This can also be expressed as --FLAG -0xC0
The FLAGs are defined in the SAM specification as follows:
Decimal
Hex
Meaning
1
0x1
template having multiple segments in sequencing
2
0x2
each segment properly aligned according to the aligner
4
0x4
segment unmapped
8
0x8
next segment in the template unmapped
16
0x10
SEQ being reverse complemented
32
0x20
SEQ of the next segment in the template being reverse complemented
64
0x40
the first segment in the template
128
0x80
the last segment in the template
256
0x100
secondary alignment
512
0x200
not passing filters, such as platform/vendor quality controls
1024
0x400
PCR or optical duplicate
2048
0x800
supplementary alignment
Filtering reads based on MAPQ
genocat --MAPQ [^]value myfile.bam.genozip
Filters lines based on the MAPQ value: INCLUDE (or EXCLUDE if value is prefixed with ^) lines with a MAPQ greater or equal to value.
Filtering non-ACTGN “bases”
genocat --bases ACGTN myfile.bam.genozip
Displays only lines in which all characters of the SEQ are one of A,C,G,T,N
genocat --bases ^ACGTN myfile.bam.genozip
Displays only lines in which NOT all characters of the SEQ are one of A,C,G,T,N
Note: In all lines missing a sequence (i.e. SEQ=*) are included in positive –bases filters (the first example above) and excluded in negative ones.
Note: The list of IUPAC chacacters can be found here: IUPAC codes
Filtering reads by species
Genozip has the unique ability to filter SAM/BAM files by species (taxonomy id). This is useful, for example, for filtering out bacterial contamination by directly removing reads that map to bacterial genomes rather than just removing reads with low mapping quality, assuming they represent contamination. See Filtering BAM or FASTQ reads by species using kraken2.
Inspecting field-level compression statistics
If optimizing the compressed file size is important, the option --stats can be used in genozip, genounzip or genocat to get a better understanding of the information content of the individual fields. For example:
$ genocat --stats myfile.bam.genozip
BAM file: myfile.bam
Alignment lines: 99,909 Dictionaries: 50 Vblocks: 2 x 16 MB Sections: 143
Genozip version: 12.0.11 conda
Date compressed: 2021-08-20 17:28:44 ACDT
License v12.0.11 granted to: ***** accepted by: ***** on 2021-07-23 14:33:51 ACDT from IP=*****
Sections (sorted by % of genozip file):
NAME GENOZIP % TXT % RATIO
QUAL 978.8 KB 38.1% 14.1 MB 45.1% 14.8X
QNAME 606.2 KB 23.6% 3.8 MB 12.0% 6.4X
SEQ 357.4 KB 13.9% 7.5 MB 23.9% 21.4X
MD:Z 131.9 KB 5.1% 598.7 KB 1.9% 4.5X
TLEN 122.1 KB 4.8% 390.3 KB 1.2% 3.2X
PNEXT 118.8 KB 4.6% 390.3 KB 1.2% 3.3X
XS:i 46.5 KB 1.8% 97.4 KB 0.3% 2.1X
CIGAR 43.8 KB 1.7% 646.3 KB 2.0% 14.7X
POS 40.1 KB 1.6% 390.3 KB 1.2% 9.7X
FLAG 29.6 KB 1.2% 195.1 KB 0.6% 6.6X
AS:i 29.3 KB 1.1% 97.4 KB 0.3% 3.3X
NM:i 23.0 KB 0.9% 97.4 KB 0.3% 4.2X
MAPQ 21.0 KB 0.8% 97.6 KB 0.3% 4.6X
TXT_HEADER 8.2 KB 0.3% 24.7 KB 0.1% 3.0X
SA:Z 5.4 KB 0.2% 13.3 KB 0.0% 2.5X
Other 2.8 KB 0.1% 1.8 MB 5.8% 666.7X
RNEXT 1.4 KB 0.1% 390.3 KB 1.2% 284.0X
XQ:i 991 B 0.0% 522 B 0.0% 0.5X
BGZF 792 B 0.0% - 0.0% 0.0X
RNAME 297 B 0.0% 390.3 KB 1.2% 1345.6X
BAM_BIN 43 B 0.0% 195.1 KB 0.6% 4646.9X
RG:Z 42 B 0.0% 195.1 KB 0.6% 4757.6X
GENOZIP vs BGZF 2.5 MB 100.0% 5.7 MB 100.0% 2.3X
GENOZIP vs TXT 2.5 MB 100.0% 31.3 MB 100.0% 12.5X
In this paritcular example, we observe that the QUAL field consumes 38.1% of the total compressed file size. Therefore, we can expect that --optimize-QUAL will significantly reduce the compressed file size. In contrast, NM:i, for example, consumes only 0.9% of the compressed file size. Therefore, we can expect that getting rid of NM:i will not significantly reduce the compressed file size.
idxstats
genocat --idxstats myfile.bam.genozip
Calculates idxstats, similar to samtools idxstats. See idxstats.
Per-contig coverage and depth
genocat --show-coverage myfile.bam.genozip
An experimental feature for calculating coverage and depth, see Coverage and Depth.
Sex assignment
genocat --show-sex myfile.bam.genozip
An experimental feature for determining the sex of a sample, see Sex assignment.
Multi-threading
By default, Genozip attempts to utilize as many cores as available. For that, it sets the number of threads to be a bit more than the number of cores (a practice known as “over-subscription”), as at any given moment some threads might be idle, waiting for a resource to become available. The --threads <number> option allows explicit specification of the number of “compute threads” to be used (in addition a small number of I/O threads is used too, usually 1 or 2).
Memory (RAM) consumption
In genozip, each compute thread is assigned a segment of the input file, known as a VBlock. By default, the size of the VBlock is set automatically to balance memory consumption and compression ratio for the particular input file, however it may be set explicitly with genozip --vblock <megabytes> (<megabytes> is an integer between 1 and 2048). A larger VBlock usually results in better compression while a smaller VBlock causes genozip to consume less RAM. The VBlock size can be observed at the top of the --stats report. genozip’s memory consumption is linear with (VBlock-size X number-of-threads).
genocat and genounzip also consume memory linearly with (VBlock-size X number-of-threads), where VBlock-size is the value used by genozip of the particular file (it cannot be modified genocat or genounzip). Usually, genocat and genounzip consume significantly less memory compared to genozip.
When using a reference file, it is loaded to memory too. If multiple genozip/ genocat / genounzip processes are running in parallel, only one copy of the reference file is loaded to memory and shared between all processes, and depending on how busy the computer is, that reference file data might persist in RAM even between consecutive runs, saving Genozip the need to load it again from disk. All this all happens behind the scenes.
