biologyguy / RD-MCL / 434

Builds Branch Commit Type Ran Committer Via Coverage
434 develop Switch to itertuples in helpers.MarkovClustering This is a huge speed increase over iterrows. Also converted self.name_order to a dict, along with a reverse lookup called self.name_order_indx push biologyguy travis-ci pending completion   set done
433 develop Convert pandas string fields into categories This gives a huge speed and memory utilization improvement. push biologyguy travis-ci pending completion   set done
432 develop Remove .append() from rdmcl.FwdScoreCorrelations._mc_rsquare_vals() Instead, instantiate a properly sized DF, then fill in the cells as needed. push biologyguy travis-ci pending completion   set done
431 develop New dev_flag --build_environment in rdmcl Allows all of the preparation dirs and files to be generated, without actually predicting any clusters. push biologyguy travis-ci pending completion   set done
430 develop New dev_flag --build_environment in rdmcl Allows all of the preparation dirs and files to be generated, without actually predicting any clusters. push biologyguy travis-ci pending completion   set done
429 develop A little refactoring and make sure concat_alignment matches whole name push biologyguy travis-ci pending completion   set done
428 develop Allow max CPUs to be specified push biologyguy travis-ci pending completion   set done
427 develop Allow the user to set the tree inference program and add arguments push biologyguy travis-ci pending completion   set done
426 develop Only use the domain-based alignment/tree inference if specified push biologyguy travis-ci pending completion   set done
425 develop Do not try to perform final placement if only a single cluster is found There's no way to make a null distribution if all sequences are in cluster. push biologyguy travis-ci pending completion   set done
424 develop (Manuscript) Implementation section push biologyguy travis-ci pending completion   set done
423 develop Rename hmm fwd score dataframe headings from rec_id[12] to seq[12] This is consistent with other dataframes, and will allow some cross interoperability. push biologyguy travis-ci pending completion   set done
422 develop Extract forward score handling from Seqs2Clusters to new class The new class is called FwdScoreCorrelations push biologyguy travis-ci pending completion   set done
421 develop Seqs2Clusters.create_hmms_for_every_rec() to mc_create_sequence_hmms() I want to run this function at the beginning of an RDMCL job, which will allow me to use FWD score similarity as an initial RBH paralogs search instead of the MSA. push biologyguy travis-ci pending completion   set done
420 develop Implement Feature based alignment/tree inference This reduces the number of characters in the alignments, but increases the confidence that those characters are homologous. Also commented out the TrimAl step. push biologyguy travis-ci pending completion   set done