X-Git-Url: http://woldlab.caltech.edu/gitweb/?p=samtools.git;a=blobdiff_plain;f=NEWS;h=41a6cc8ebe05185ea8e97f0639fede7ce15675d9;hp=a600bb1d2519fb4a139a406f223da55dddcc1b80;hb=58ef643243f8e017d859c9fb27ba8a5f3f4517c0;hpb=8a8212c76299ff302cdd41be704d8e0308bd2b1b

diff --git a/NEWS b/NEWS
index a600bb1..41a6cc8 100644
--- a/NEWS
+++ b/NEWS
@@ -1,3 +1,74 @@
+Beta Release 0.1.18 (2 September, 2011)
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+Notable changes in samtools:
+
+ * Support the new =/X CIGAR operators (by Peter Cock).
+
+ * Allow to subsample BAM while keeping the pairing intact (view -s).
+
+ * Implemented variant distance bias as a new filter (by Petr Danecek).
+
+ * Bugfix: huge memory usage during indexing
+
+ * Bugfix: use of uninitialized variable in mpileup (rare)
+
+ * Bugfix: wrong BAQ probability (rare)
+
+Notable changes in bcftools:
+
+ * Support indel in the contrast caller.
+
+ * Bugfix: LRT2=nan in rare cases
+
+(0.1.18: 2 September 2011, r982:295)
+
+
+
+Beta Release 0.1.17 (6 July, 2011)
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+With the maturity of `mpileup' and the lack of update in the `pileup' command,
+the `pileup' command is now formally dropped. Most of the pileup functionality,
+such as outputting mapping quality and read positions, have been added
+`mpileup'.
+
+Since this release, `bcftools view' is able to perform contrast SNP calling
+(option -T) for discovering de novo and/or somatic mutations between a pair of
+samples or in a family trio. Potential mutations are scored by a log likelihood
+ratio, which is very simple in math, but should be comparable to more
+sophisticated methods. Note that getting the score is only the very first step.
+A lot more need to be done to reduce systematical errors due to mapping and
+reference errors and structural variations.
+
+Other notable changes in samtools:
+
+ * Improved sorting order checking during indexing.
+
+ * Improved region parsing. Colons in reference sequence names are parsed
+   properly.
+
+ * Fixed an issue where mpileup does not apply BAQ for the first few reads when
+   a region is specified.
+
+ * Fixed an issue where `faidx' does not work with FASTA files with long lines.
+
+ * Bugfix: wrong SP genotype information in the BCF output.
+
+Other notable changes in bcftools:
+
+ * Output the ML esitmate of the allele count.
+
+ * Added the HWE plus F<0 filter to varFilter. For multiple samples, it
+   effectively filters false heterozygous calls around centromeres.
+
+ * For association mapping, perform both 1-degree and 2-degree test. The
+   2-degree test is conservative but more robust to HWE violation.
+
+(0.1.17: 6 July 2011, r973:277)
+
+
+
 Beta Release 0.1.16 (21 April, 2011)
 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~