Closed colindaven closed 7 years ago
Dear Colin,
That's true, the maximum value for K is 256, This limitation is posed by KMC2, the kmer counting tool we use for building the kmer index. Actually, it is much larger than what we need for building a pangenome. Choosing larger valued for K decrease the connectivity of the pangenome and you would not detect the shared sequences smaller than K. In the paper you can see that even for 1000 human genomes K=48 suffices.
Cheers,
Siavash
Siavash Sheikhizadeh Anari Bioinformatics Group Wageningen University Droevendaalsesteeg 1, 6708PB Building/Room: 107/W1.Ad.054
From: colindaven notifications@github.com Sent: 16 September 2016 16:27 To: Sheikhizadeh/pantools Subject: [Sheikhizadeh/pantools] kmer 257 not possible, 256 is max (#2)
This command throws the following error. K=256 works.
java -Xmx450g -jar /home/bioinformatics/NAS01/programs/pantools/pantools/dist/pantools.jar build 257 ./db input_fasta.fofn
------------------------------- PanTools ------------------------------- K should be between 0 and 257 !
You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHubhttps://github.com/Sheikhizadeh/pantools/issues/2, or mute the threadhttps://github.com/notifications/unsubscribe-auth/AJbc0CTUYUGW_mv5gxTWPFzMCmofNGS3ks5qqqdsgaJpZM4J-_zr.
This command throws the following error. K=256 works.
java -Xmx450g -jar /home/bioinformatics/NAS01/programs/pantools/pantools/dist/pantools.jar build 257 ./db input_fasta.fofn
------------------------------- PanTools ------------------------------- K should be between 0 and 257 !