Somaticseq

When and only when SomaticSeq.Wrapper.sh is not in training or predicting mode, it will fall back into a naive consensus mode:
- A call agreed upon by over 50% of the callers will be labeled PASS
- A call agreed upon by at least 1/3, but no more than 50% of the callers, will be labeled LowQual.
- The remaining will be labeled REJECT.

Caller consensus is only invoked if there is no trained classifier specified or ground truth file to train a classifier.

- Extract information from BAM files using Pysam
- Reduced some features that are redundant, but added some features from the paper.
- The tsv2vcf.py script has added flexibility and configurability.

Use -C 50 and -q 1 options in mpileup in the Wrapper script.

- SSeq_merged.vcf2tsv.py can now accept pileup files to extract read depth and DP4 (reference forward, reference reverse, alternate forward, and alternate reverse) information (mainly for indels). Previously, that information can only be extracted from SAMtools VCF. Since the SAMtools or HaplotypeCaller generated VCFs hardly contain any indel information, this option improves the indel model. The SomaticSeq.Wrapper.sh script is modified accordingly.
- Extract mapping quality (MQ) from VarDict output if this information cannot be found in SAMtools VCF (also mostly benefits the indel model).
- Indel length now positive for insertions and negative for deletions, instead of using the absolute value previously.

Automate both model training and mutation prediction. Training or prediction mode based on files supplied. No significant change from previous version.

This version was used for the results in the SomaticSeq manuscript.