Version 1.0.3
Fix the database path in the `run_diamond` fuction (issue #128)

Version 1.0.2
Add download retry logic and update dependencies (issue #108)

Version 1.0.1
* Program checks for "prodigal-gv" instead of "prodigal" (issue #92)

Version 1.0.0
* Fixed problem parallelizing prodigal (issue #88)
* Fixed problem with logging function (issue #82)
* Improved DTR detection; continues checking for DTRs after the first substring match fails
* Switched to prodigal-gv to accomodate alternative codes (https://github.com/apcamargo/prodigal-gv)
* Count cannonical kmers for calculaing kmer_freq statistic (includes reverse complement)
* Fixed support for compressed inputs

Version 0.9.0
* Check for DIAMOND errors and DIAMOND output file (issues #15 and #39); was NOT fixed in v0.8.1
* DIAMOND database built on the fly to avoid incompatabilities (issue #57)
* Minor fixes (issues #51, #47)

Version 0.8.1
* Check for DIAMOND errors and DIAMOND output file (issues #15 and #39)

Version 0.8.0
* Check if input FASTA file is empty, wrongly formatted, or duplicated sequence IDs (issue #14, pull #5)
* Check whether the execution of hmmsearch was completed before skipping it (pull #5)
* Speed up kmer based genome copy number estimation (issue #26, pull #4)
* Add a --remove_tmp flag to quality_summary and end_to end modules (pull #3)

Version 0.7.0
* update to contamination module to reduce false positive proviruses
 -require at least 2 host genes in proviral host regions, except for contigs with <= 10 genes. reduces false positives in real data.
 -improved sliding window algorithm: trailing window grows from last breakpoint while leading window has fixed size
 -size of leading window is dynamically set: 30% of genes with min of 15 and max of 50
 -additional statistics reported for each host/viral region
* update to hmm-based completeness for more robust estimates
 -improved identification of 5% and 95% quantiles
 -use all HMMs to calculate completeness: each yields the 5%-95% completeness range, and these ranges are averaged after weighting each by the HMMs coefficient of variation (based on how variable reference genome sizes are for the HMM)
 -improves concordance with AAI-based completeness
* update to complete_genomes module (formerly called repeats)
 -added confidence levels, based on comparison to completeness and detection of artifacts

Version 0.6.0
* Bug fix in completeness module that suppressed HMM-based results
* HMM-based completeness now reported as range: 90% confidence interval = X-Y% complete
* Warnings added to quality_summary.tsv for range of conditions
* Minor parameter change for genome copy number estimation
* Terminal repeat sequence included in output file (e.g. GCACTTC...)
* Added terminal repeat metadata: ambig_freq and max_base_freq
* Dropped automated filtering of terminal repeats: up to user to decide how to handle flagged seqs
* Quality summary reports multiple types of termini when present (e.g. DTR & ITR on same contig)
* cleaned_contigs.fna split into viruses.fna and proviruses.fna
* Renamed field 'prophage' to 'provirus' in quality_summary.tsv

Version 0.5.1
* Added download_database module to automatically download latest version of db
* Added update_database module to update db with new genomes
* Added end_to_end module to run all CheckV modules
* Added documentation for output files in README
* Help text reformatted

Version 0.4.0
* New HMM-based completeness estimation
* New estimation of genome copy #
* Completeness capped at 100% in quality_summary.tsv output
* Dropped 'comments' field in quality_summary.tsv output
* Computation of aligned fraction fixed
* terminal_repeats module renamed to repeats (to accomonate repeat based estimation of genome copy #)
* All TRs now reported in repeats.tsv
* New field 'repeat_flagged' which indicates low complexity/repetetive repeats
* Added RELEASE_NOTES.txt file
* CheckV print program and database versions
* CheckV checks for input files in quality_summary module

Version 0.3.0
* Added documentation to README.md
* Support for compressed input files (gz, bz2, xz)
* Param change for contamination module (win size: 35 to 40, min % host genes 20% to 30%)
* Generate cleaned_contigs.fna
