zsutx2005 / rna-bloom Goto Github PK

RNA-Bloom is a fast and memory-efficient de novo transcript sequence assembler for bulk and single-cell paired-end RNA-seq data.

Written by Ka Ming Nip 📧

©️ 2018 Canada's Michael Smith Genome Sciences Centre, BC Cancer

• Java SE Runtime Environment (JRE) 8

Check your Java version:

java -version

Example:

java version "1.8.0_101"
Java(TM) SE Runtime Environment (build 1.8.0_101-b13)
Java HotSpot(TM) 64-Bit Server VM (build 25.101-b13, mixed mode)

1. Download the binary tarball rnabloom_vX.X.X.tar.gz from the releases section.
2. Extract the downloaded tarball with the command:

tar -zxf rnabloom_vX.X.X.tar.gz

3. RNA-Bloom is ready to use, ie. java -jar /path/to/RNA-Bloom.jar ...

There is nothing to compile/configure/build! 👍

⚠️ RNA-Bloom only supports paired-end RNA-seq data. Input reads must be in either FASTQ or FASTA format and may be compressed with GZIP.

java -jar RNA-Bloom.jar -left LEFT.fastq.gz -right RIGHT.fastq.gz -revcomp-right -t THREADS -outdir OUTDIR

java -jar RNA-Bloom.jar -stranded -left LEFT.fastq.gz -right RIGHT.fastq.gz -revcomp-right -t THREADS -outdir OUTDIR

Note that dUTP protocols produce reads in the F2R1 orientation, where /2 denotes left reads in forward orientation and /1 denotes right reads in reverse orientation. In this case, please specify your reads paths as -left reads_2.fastq -right reads_1.fastq.

java -jar RNA-Bloom.jar -pool READSLIST.txt -revcomp-right -t THREADS -outdir OUTDIR

cell1 /path/to/cell1/left.fastq.gz /path/to/cell1/right.fastq.gz
cell2 /path/to/cell2/left.fastq.gz /path/to/cell2/right.fastq.gz
cell3 /path/to/cell3/left.fastq.gz /path/to/cell3/right.fastq.gz

Columns are separated by space/tab characters.

This file consists of 3 columns, ie.

1. cell ID
2. path of left reads
3. path of right reads

java -jar RNA-Bloom.jar -fpr 0.1 -nk 28077715 ...

The number of unique k-mers in your dataset can be estimated efficiently with ntCard.

When running ntCard, please specifiy the same k-mer size to be used in RNA-Bloom (eg. 25), eg.

ntcard -k 25 -c 65535 -p outdir/freq LEFT.fastq.gz RIGHT.fastq.gz

ntCard would generate a histogram file outdir/freq_k25.hist, where F0 on the 2nd row is the number of unique k-mers, eg.

F1	140110302
F0	28077715

Alternatively, you can use the -ntcard option in RNA-Bloom if ntcard is already in your PATH, eg.

java -jar RNA-Bloom.jar -fpr 0.05 -ntcard ...

java -jar RNA-Bloom.jar -mem 3 ...

Otherwise, it is adjusted automatically based on the size of input read files.

java -jar RNA-Bloom.jar -help

java -Xmx1g -jar RNA-Bloom.jar ...

This option does not need to be set larger than the total Bloom filter size.

Other JVM options may also be used.