redlin810/assemblyline

Automatically exported from code.google.com/p/assemblyline

assemblyline's Issues

Assigning strand to unstranded transfrags

Unstranded transfrags in intronic regions should be assigned strand 
corresponding to the intron orientation unless higher priority information is 
available from strand-specific libraries

Otherwise unstranded transfrags will not get culled out using the intron 
filters.

Potentially could mark intronic transcripts during initial parsing.

Original issue reported on code.google.com by [email protected] on 26 Mar 2013 at 7:52

Incorporate poly-adenylation sites into assembler

Using poly-A site lists will improve definition of transcript ends and enable 
transcripts to be split correctly

1. Discover poly-A sites using RNA-Seq, ChIP-Seq, etc
2. Add to known DB of poly-A sites
3. Incorporate poly-A DB into assembly step

Original issue reported on code.google.com by [email protected] on 9 Dec 2011 at 5:11

Improve bias smoothing algorithm

Study and improve least-squares smoothing algorithm for node densities

Original issue reported on code.google.com by [email protected] on 9 Dec 2011 at 5:13

Use expression signal rather than number of transcripts when computing prior weights

Use expression signal rather than number of transcripts when computing prior 
weights

Original issue reported on code.google.com by [email protected] on 27 Mar 2013 at 2:52

Maintain intron scores

For each intron maintain scores

Makes it easier to filter intronic transfrags

use interval trees do intersect intronic transfrags with introns and find 
highest scoring overlapping intron to use for filter

Original issue reported on code.google.com by [email protected] on 26 Mar 2013 at 7:54