xinglab/seastar

SEASTAR - Systematic Evaluation of Alternative transcription STArt site in RNA

seastar's Issues

rMATS fails in test run

I have recently downloaded SEASTAR following instructions. When running the test script, the last part of the pipeline, rMATS, fails. I would like to know how to solve the problem, because I would really like to use the tool you wrote. Attached, you can find the output of the test run and, at then end, the list of packages installed in my conda environment.
output_test_1.txt
Thanks a lot for your support

BAM Record Error

I am trying to use SEASTAR on human RNASeq data. My inputs are BAM files aligned with STAR. They are single-end, reverse strand reads (sequenced on Illumina HiSeq 2500). Whenever I try to run "SEASTAR.sh", I get an error message that says "BAM record error: found spliced alignment without XS attribute" when SEASTAR is at the "Process locus: " stage. I am using the following command:

bash /home/travisblimkie/bin/SEASTAR-master/SEASTAR.sh
-A /mnt/analysis1/myproj/STAR_files/sample1_Aligned.sortedByCoord.out.bam
-B /mnt/analysis1/myproj/STAR_files/sample4_Aligned.sortedByCoord.out.bam
-o /mnt/analysis1/myproj/Seastar_runs/sample1
-g /mnt/analysis1/Genomes/Human_GRCh38_bowtie_index/Homo_sapiens.GRCh38.93.gtf
-i /mnt/analysis1/Genomes/Human_GRCh38_bowtie_index/human.chrom.sizes
-s /mnt/analysis1/Genomes/Human_GRCh38_bowtie_index/Homo_sapiens.GRCh38.dna.primary_assembly.fa
-S s

From a quick search online, it seems like it has something to do with incorrect parameters being given to cufflinks. Any suggestions would be greatly appreciated.

Thanks,
Travis

The meaning of length, IncFormLen, and SkipFormLen in rMATS_Result.txt

ID	class_code	gene_id	gene_short_name	locus	length	tss_chr	tss_ss	tss_es	strand	IC_SAMPLE_1	SC_SAMPLE_1	IC_SAMPLE_2	SC_SAMPLE_2	IncFormLen	SkipFormLen	PValue	FDR	IncLevel1	IncLevel2	IncLevelDifference
TSS331	-	ENSG00000048052	HDAC9	chr7:18535368-19036993	4659	chr7	18535333	18535477	+	0,0	10,18	39,17	37,17	145	145	0.00017	0.00424	0.0,0.0	0.513,0.5	-0.506
TSS325	-	ENSG00000048052	HDAC9	chr7:18126571-18706546	2524	chr7	18126572	18126836	+	8,16	11,17	20,10	118,52	265	265	0.003994	0.049931	0.421,0.485	0.145,0.161	0.3
TSS256	-	ENSG00000229091	HSPA8P8	chr7:10490937-10492153	1216	chr7	10490938	10492153	+	79,102	0,0	45,104	0,0	1216	1216	1	1	1.0,1.0	1.0,1.0	0
TSS261	-	ENSG00000106443	PHF14	chr7:11013498-11143282	4276	chr7	11013499	11013951	+	87,23	0,0	51,6	0,0	453	453	1	1	1.0,1.0	1.0,1.0	0
TSS275	-	ENSG00000230435	AC004160.4	chr7:11446581-11453705	563	chr7	11446582	11446793	+	71,34	0,0	23,13	0,0	212	212	1	1	1.0,1.0	1.0,1.0	0
TSS276	-	ENSG00000106460	TMEM106B	chr7:12250925-12276886	6432	chr7	12250867	12251051	+	11,9	0,0	27,4	0,0	185	185	1	1	1.0,1.0	1.0,1.0	0
TSS279	-	ENSG00000236048	AC013470.6	chr7:12510977-12512171	1194	chr7	12510978	12512171	+	42,11	0,0	7,6	0,0	1194	1194	1	1	1.0,1.0	1.0,1.0	0
TSS293	-	ENSG00000122644	ARL4A	chr7:12726480-12728804	1296	chr7	12726481	12726668	+	17,2	0,0	58,36	0,0	188	188	1	1	1.0,1.0	1.0,1.0	0

This is the first few lines of the testrun output, according to the document, it seems the length strands for the chromosomal position of the TSS. However, these value are far more smaller than the tss_ss, which seems quite unreasonable. The Next problem is about IncFormLen and SkipFormLen. Why the two value are identical. The two exon length should be different.