ezLncPred: A Python package with integrated lncRNA prediction methods

ezLncPred is an comprehensive python package for LncRNA identification which integrates 9 state-of-the-art lncRNA prediction models. ezLncPred python package provides a convenient command line method for researchers who intends to identify lncRNAs. The ezLncPred web server tool provides a straight-to-the-point answer for input fasta file, or an email for time-consuming identification task.

Integration

ezLncPred currently provides 9 LncRNA prediction models, which are listed as follows.

• CNCI
• CPC2
• lgc
• PLEK
• CPAT
• CPPred
• longdist
• PredLnc-GFStack
• LncADeep

Web server

Python package installation

• Prerequisite
  • python 3.0 version (or above)
  • linux operating system
  • C/C++ compiler(for PLEK)
• Download ezLncPred by

pip install ezLncPred

Help

For detailed message of ezLncPred, run

ezLncPred --manual

For detailed message of each model and their parameters, run

ezLncPred  --manual [model]{CNCI,CPC2,CPAT,lgc,CPPred,GFStack,longdist,PLEK,LncADeep}

Usage

ezLncPred offers a total of 9 LncRNA prediction models, each with a different variety of parameter choices, users can refer to the list below to customarize your prediction procedure. First, ezLncPred must receive at least three parameters to specify the input file output directory and prediction model

-i --input		fasta format input files

-o --output		the output directory to store the identification results

-m --manual		show manuals

-v --version		show program's version number and exit

For example, run

ezLncPred -i your_fasta_file -o output_directory model [parameters]

Individual model parameters

CNCI

-h --help		show this help message and exit

--parallel		assign the running CUP numbers

-p {ve,pl} --species {ve,pl}
			assign the classification models ("ve" for vertebrate species, "pl" for plat species)

example

	ezLncPred CNCI -h
	ezLncPred -i example.fa -o results CNCI
	ezLncPred -i example.fa -o results CNCI --parallel
	ezLncPred -i example.fa -o results CNCI -p ve

CPC2

-h --help		show this help message and exit

-r REVERSE --reverse
			REVERSE also check the reverse strand

example

	ezLncPred CPC2 -h
	ezLncPred -i example.fa -o results CPC2
	ezLncPred -i example.fa -o results CPC2 -r REVERSE

lgc

-h --help		show this help message and exit

example

	ezLncPred lgc -h
	ezLncPred -i example.fa -o results lgc

PLEK

-h --help		show this help message and exit

--thread		the number of threads to run the PLEK programme

--isoutmsg		Output messages to stdout(screen) or not. "0" means 
			that PLEK be run quietly. "1" means that PLEK outputs
			the details of processing. Default value: 0
				
--isrmtempfile		Remove temporary files or not. "0" means that PLEK 
			retains temporary files. "1" means that PLEK remove 
			temporary files. Default value: 1

example

	ezLncPred PLEK -h
	ezLncPred -i example.fa -o results PLEK
	ezLncPred -i example.fa -o results PLEK --thread 4
	ezLncPred -i example.fa -o results PLEK --isoutmsg 1
	ezLncPred -i example.fa -o results PLEK --isrmtempfile 0

CPAT

-h --help		show this help message and exit

-p --species    	{Human,Mouse,Fly,Zebrafish}
			specify the species of the LncRNAs choose from Human 
			Mouse Fly Zebrafish (note that the first character 
			is upper case)
				
-s --start		Start codon (DNA sequence, so use 'T' instead of 'U')
			used to define open reading frame (ORF), default is ATG
				
-t --stop		Stop codon (DNA sequence, so use 'T' instead of 'U')
			used to define open reading frame (ORF). Multiple stop
			codons should be separated by ',' default is TAG,TAA,TGA

example

	ezLncPred CPAT -h
	ezLncPred -i example.fa -o results CPAT
	ezLncPred -i example.fa -o results CPAT -p Human
	ezLncPred -i example.fa -o results CPAT -s TAG
	ezLncPred -i example.fa -o results CPAT -t ATG,TGA,TTA

CPPred

-h --help		show this help message and exit

-p --species		{Human,Integrated}
			the model of the species to choose (Human,Integrated).

example

	ezLncPred CPPred -h
	ezLncPred -i example.fa -o results CPPred
	ezLncPred -i example.fa -o results CPPred -p Integrated

longdist

-h --help		show this help message and exit

-z <200>, --size <200>
			Minimun sequence size to consider. Default is 200.

-p --species	{Human,Mouse}
			the model of the species to choose (human,mouse).

example

	ezLncPred longdist -h
	ezLncPred -i example.fa -o results longdist
	ezLncPred -i example.fa -o results longdist -z 150
	ezLncPred -i example.fa -o results longdist -p Human

PredLnc-GFStack

-h --help		show this help message and exit

-p --species		{human,mouse}
			choose a species type from Human and Mouse

example

	ezLncPred GFStack -h
	ezLncPred -i example.fa -o results GFStack
	ezLncPred -i example.fa -o results GFStack -p human

LncADeep

-h --help		show this help message and exit

-mt --modeltype	{full,partial}
			the model used for lncRNA identification,
			choose from partial full default is partial
			default is "partial".
				
-HMM --HMMthread
			the thread number of using HMMER, default is 8
				
-p --species	{human,mouse}
			the model of the species to choose (human,mouse).
			default is "human".

-t --thread	THREAD
                		The number of threads for running the LncADeep program.default is 8.

example

	ezLncPred LncADeep -h
	ezLncPred -i example.fa -o results LncADeep
	ezLncPred -i example.fa -o results LncADeep -mt full
	ezLncPred -i example.fa -o results LncADeep -HMM 4
	ezLncPred -i example.fa -o results LncADeep -p human
	ezLncPred -i example.fa -o results LncADeep -t 4

Test case

We used lncRNA fasta file (1.0M) to test the cost time of prediction by each model. We run the default command without any additional parameters on Intel(R) Xeon(R) Gold 6146 CPU @ 3.20GHz. The test results are for reference only, and the specific values ​​depend on the specific situation.

Model	Real Time	User Time	Sys Time
CNCI	2m14.578s	2m13.992s	0m0.335s
CPC2	0m1.928s	0m1.756s	0m0.163s
CPAT	0m2.076s	0m1.896s	0m0.182s
lgc	0m1.640s	0m1.476s	0m0.162s
CPPred	0m13.973s	0m13.767s	0m0.200s
GFStack	0m44.763s	4m35.646s	0m14.584s
longdist	0m2.689s	0m2.569s	0m0.119s
PLEK	0m21.603s	1m30.905s	0m1.431s
LncADeep	4m1.667s	6m12.511s	0m23.426s

Papers

• CNCI : “Utilizing sequence intrinsic composition to classify protein-coding and long non-coding transcripts”, Sun et al. (2013).
• CPC2 : “CPC2: a fast and accurate coding potential calculator based on sequence intrinsic features”, Kang, J., et al. (2017).
• CPAT : “CPAT: Coding-Potential Assessment Tool using an alignment-free logistic regression model”, Wang et al. (2013).
• lgc : “Characterization and identification of long noncoding RNAs based on feature relationship”, Wang et al. (2019).
• CPPred : “CPPred: coding potential prediction based on the global description of RNA sequence”, Tong et al. (2019).
• GFStack : “PredLnc-GFStack: A Global Sequence Feature Based on a Stacked Ensemble Learning Method for Predicting lncRNAs from Transcripts”, Liu et al. (2019).
• longdist : “A Support Vector Machine based method to distinguish long non-coding RNAs from protein coding transcripts”, W.Schneider, H., et al. (2017).
• PLEK : “PLEK: a tool for predicting long non-coding RNAs and messenger RNAs based on an improved k-mer scheme”, Li, A., et al. (2014).
• LncADeep : “LncADeep: An ab initio lncRNA identification and annotation tool based on deep learning”, Yang et al. (2017).