I ran into error message at jCounts step. It seems related to mate pair on different chromosomes. If I filter out bam files for mate pair sitting on different chromosomes (filed 7 with = sign means mate pair on same chromosome): samtools view -h input.bam | awk '($7 == "=")' > output.sam, then it can run.

txdb<-makeTxDbFromEnsembl(organism="Homo sapiens", release=103, circ_seqs=NULL, server="ensembldb.ensembl.org", username="anonymous", password=NULL, port=0L, tx_attrib=NULL)

features<-binGenome(txdb)

gbcounts <- gbCounts( features = features, targets = samples, minReadLength = 150, maxISize = 50000,
libType="PE", strandMode=0)

asd<- jCounts(counts = gbcounts,features = features,minReadLength = 150,libType="PE",strandMode=0,threshold = 5,minAnchor = 10)

# Error: Error in .local(x, use.names, use.mcols, ...) : For some pairs in 'x', the 2 alignments are not on the same chromosome. Cannot associate a unique genomic range to such # pairs. Please call granges() with 'on.discordant.seqnames="drop"' to drop these pairs, or with 'on.discordant.seqnames="split"' to represent each of them with 2 genomic ranges # in the returned GRanges object. Note that in both cases the returned object won't be parallel to 'x'. Alternatively, please consider using grglist() instead of granges() to turn 'x' 
 # into a GRangesList object instead of a GRanges object. See ?GAlignmentPairs for more information.
#  

sessionInfo( )

Hi Stephen, thanks for using ASpli

How many reads do you discard?

You can control this issue at the alignment step in order to avoid them.

Or you can treat them as SE at jCounts step