SL discovery

🚀 Introduction

SLseek is a bioinformatics pipeline for discovering highly represented sequences at the ends of transcripts, usually spliced leader (SL) sequences in organisms with trans-splicing. It analyses full-length transcriptomic data using a k-mer-based approach, avoiding costly steps such as mapping or transcriptome assembly, and does not require a reference genome.

This pipeline is built using Nextflow, a scalable workflow management tool.

🔁 Pipeline summary

1. Extract ends of transcripts from the input full-length transcripts;
2. Count kmers using either FastK and/or Jellyfish;
3. Get Putative SL sequences (e.g, sequences composed of kmers highly present at the ends of transcripts, within a specified size and mostly non-repetitive);
4. Cluster Putative SLs using CD-HIT.

Optionally, the pipeline can generate a k-mer histogram and a heatmap plot with putative SL assessment metrics.

---

⚙️ Usage

Note

Please refer to this page to install and set-up Nextflow. You can test the installation following this link with -profile test.

nextflow run transsplicing.nf -profile [standard | lsf [ ,singularity ]] \
    --outdir <path/to/outdir/> \
    --transcripts <path/to/transcripts.fasta> \
    --kmer_counting_tool (fastk | jellyfish) \
    -c [transsplicing.config] \
    [ -resume ]

---

📖 Understanding parameters

Parameter	Default	Required	Description	Options/Input Type
outdir		Yes	A path to where saving outputs.	string
transcripts		Yes	A fasta file with full-length transcripts	string
length	100	No	Length of extracted sequences for analysis - Should encompass SL size.	int
k	21	No	K-mer size for analysis - Should be less than SL-size and typically an odd number.	int
lower_cov_thrsld_counting	150	No	Minimum coverage of kmers to filter using Jellyfish or FastK	int
lower_cov_thrsld_extracting	150	No	Minimum coverage of kmers to filter in k-mer analysis steps	int
max_distance	3	No	Minimum distance to consider a K-mer within the same group	int
extra_border	0	No	Number of extranucleotides flanking putative SLs from K-mers.	int
size_limit_max	30	No	Maximum size of a Putative SL.	int
size_limit_min	20	No	Miiumum size of a Putative SL.	int
entropy_lim	1.40	No	Minumum entropy value for Putative SL to avoid repetitive sequences	float
nucleotide_limit	0.7	No	Percentage of nucleotides to consider a kmer as uninformative. E.g., for 0.7, kmers with 70% of the same nucleotides will be removed	float
identity_thrsld	0.90	No	Identity threshold for CD-HIT Putative SL clustering.	float
heatmap	false	No	If called or true, will generate a heatmap plot with putative SL analysis metrics	boolean (E.g., true; false)
kmerplot	false	No	If called or true, will generate a kmer histogram with kmer counting process	boolean (E.g., true; false)
kmer_counting_tool	jellyfish	Yes	kmer counting tool	string (E.g., jellyfish or fastk)

---

📦 Using the SLseek Singularity Container

The SLseek workflow can use a Singularity image for easy reproducibility. You can pull it from the Sylabs Cloud library and run it directly.

1. Pull the container

singularity pull library://beatriz-estevam/bio/slseek:1.0.0

This will download slseek.sif to your current directory. SLseek expects it in the containers directory.

2. Use at least the --profile singularity to run the pipeline with it

nextflow run transsplicing.nf -c transsplicing.config -profile singularity

---

🧪 Test data

A set of data to test is provided within the project repository to help users quickly verify the installation and functionality of the pipeline. You can find all relevant files in the dedicated test/ directory of this repository.

Running tests

! Assuming you are in the root directory of the pipeline

nextflow run transsplicing.nf \
        -c transsplicing.config
        --transcripts test/celegans_slidr_transcripts.fa \
        --k 17 \
        --outdir "test/results/" \
        --kmer_counting_tool "fastk" \
        --heatmap --kmerplot \
        -profile lsf, singularity \
        -resume 

RESULTS

Kmer plot	heatmap

SL_ID	putative_sequence	entropy	avg_coverage	supporting_kmers	#_transcripts_in_cluster
putativeSL.56	CGGCAATTTGCCGATTTGCCGGAAATTTTC	1.97248976687495	295.8	15	258
putativeSL.84	CGGAATTGAAAATTTCCGGCAAATCGGCAA	1.94939785333961	285.266666666667	15	116
putativeSL.294	CATTTTCTCAAACTTGGGTAATTAAACCCC	1.89010748014126	1777.5	14	2825
putativeSL.521	AATTGAAAATTTCCGGCAAATCGGCAAATT	1.89892956369094	300.666666666667	15	5
putativeSL.786	GGAATGGAATGGAATGGAATGGAATGGAAT	1.52192809488736	188.6	15	53
putativeSL.1454	AAATTGCCGGAATTGAAAATTTCCGGCAAA	1.91387559754782	217.2	15	110
putativeSL.1934	GCAAATCGGCAAATTGCCGGAATTGAAAAT	1.91387559754782	204.142857142857	14	119
putativeSL.2378	TCAATTCCGGCAATTTGCCGATTTGCCGGA	1.98387111321815	231.357142857143	14	40
putativeSL.3495	GAAAATTTCCGGCAAATCGGCAAACCGGCA	1.9167161869614	288.75	12	68
putativeSL.1294	CGATTTGCCGGAAATTTTCAATTCCGGCA	1.9841165493995	229.5	14	1
putativeSL.338	TTACCTCAAACTTGGGTAATTAAACCCT	1.89788985125302	1768.23076923077	13	367
putativeSL.2328	CGGAAATTTTCAATTCCGGCAATTTGCC	1.97014297933532	211.230769230769	13	5
putativeSL.4202	GAGGGTTTAATTACCCAAGTTTGAGAAT	1.89788985125302	1893.53846153846	13	2
putativeSL.288	CATCTCAAACTTGGGTAATTAAACCT	1.9006709673922	2069.81818181818	11	52
putativeSL.7	CGGTTTGATCTACAAAAAATGCGGG	1.95740581142552	245.3	10	244
putativeSL.10	CATATTTGACGCGCAAAATATC	1.9196905740497	162.666666666667	6	113
putativeSL.6	GGACCACCAGGACCACCAGG	1.57095059445467	185.2	5	143

nextflow run transsplicing.nf \
        -c transsplicing.config
        --transcripts test/celegans_slidr_transcripts.fa \
        --lower_cov_thrsld_counting 500 \
        --k 17 \
        --outdir "test/results/" \
        --kmer_counting_tool "jellyfish" \
        --heatmap --kmerplot \
        -profile lsf, singularity \
        -resume 

Kmer plot	heatmap

SL_ID	putative_sequence	entropy	avg_coverage	supporting_kmers	#_transcripts_in_cluster
putativeSL.31	GGGGTTTAATTACCCAAGTTTGAGA	1.92154193333593	2376.1	10	3174

nextflow run transsplicing.nf \
        -c transsplicing.config
        --transcripts test/celegans_slidr_transcripts.fa \
        --lower_cov_thrsld_counting 500 \
        --k 21 \
        --outdir "test/results/" \
        --kmer_counting_tool "fastk" \
        --heatmap --kmerplot \
        -profile lsf, singularity \
        -resume 

Kmer plot	heatmap

SL_ID	putative_sequence	entropy	avg_coverage	supporting_kmers	#_transcripts_in_cluster
putativeSL.31	GGGGTTTAATTACCCAAGTTTGAGA	1.92154193333593	1760.83333333333	6	3174