Update README.md

Ensured workers close after parallel processing\n Fixed an issue where the worker created during parallel processing in AlignmentConverter failed to close until the pipeline finishes running on every dataset. Further, allowed AlignmentPipeline to use AlignmentConverter in parallel.

Added more details on AlignmentConverter usage

Removed unnecessary import statement: psutil

Updated user guide.

Updated user guide

Fixed improper merge

Merged the updated AlignmentConverter.py to master. \nThe update 1) removes conflicts with newer versions of numpy and 2) permits running the function in parallel.

Update README.md

Update README.md

Updated Readme

Resolved a deprecation issue with later versions of numpy. \n When updating the MD string, the array of nucleotides reflecting the span of the consensus to which the read mapped was one element longer than it should have been. This is sliced by an array of bools defined by the CIGAR string, and this boolean array is the correct length. In early versions of Numpy, when an array is sliced by a shorter bool array, the extra elements in the sliced array are simply excluded, meaning the extra nucleotide has no impact on behavior of the function. However,in later versions of Numpy (some version >1.10.4) this a mismatch in the length between an array and the bool array used to slice it raises an array. \n I removed the extra nucleotide from the array. I also replaced several functions with numpy.fromstring to avoid a different potential deprecation.

Updated manual to describe how to summarize copy number for repeat families.

Revert "Updated User Guide"

Updated User Guide