forked from bigdatagenomics/adam
-
Notifications
You must be signed in to change notification settings - Fork 1
/
CHANGES.txt
33 lines (22 loc) · 1.63 KB
/
CHANGES.txt
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
ADAM Changelog
Trunk (not yet released)
NEW FEATURES
* Added ability to load and merge multiple ADAM files into a single RDD.
* Pairwise, quantitative ADAM file comparisons: the CompareAdam command has been extended to calculate
metrics on pairs of ADAM files which contain the same reads processed in two different ways (e.g.
two different implementations of a pre-processing pipeline). This can be used to compare different
pipelines based on their read-by-read concordance across a number of fields: position, alignment,
mapping and base quality scores, and can be extended to support new metrics or aggregations.
* Added FASTA import, and RDD convenience functions for remapping contig IDs. This allows for reference
sequences to be imported into an efficient record where bases are stored as a list of enums. Additionally,
convenience values are calculated. This feature was introduced in PR #79.
* Added helper functions for properly generating VCF headers for VCF export. This streamlines the process
of converting ADAM Variant Calls to the legacy VCF format. This was added in PR#85.
OPTIMIZATIONS
IMPROVEMENTS
* ISSUE 92: improved the representation of the types of 'optional' fields from the BAM, and their encoding
in the 'attributes' field of ADAMRecord. This encoding now includes the type, and should no longer be
lossy, therefore making it possible to write code to re-export a BAM from the ADAM file in the future.
BUG FIXES
* Fixed issues where VCF header was not being written correctly. This prevented variant calls from being
written after conversion. This was fixed in PR#85.