dsoldevila / qmprimers Goto Github PK

Mismatch location (mismXX) starts at 1 in the csv output, but the primer's location (X_pos) not (I think)

In get_csv() from Alignment class

Primers should be able to start before the dna sequence.

Save results while matching in order to not lose everything if there is an unexpected crash.

If max amplicon makes the primer pair larger than the dna sequence, but not the min amplicon, make max amplicon smaller instead of skipping.

guardar en el nend en un arixu a part amb els seus stats corresponent.

Verificar que el random sample funciona bé.
Comprovar

Length updating wrong when adding more than one element at once

Implement multiprocessing to speed up calculations

Code first approach of the simulation step.

Make Biopython pairwise2 return all results with acceptable missmatches instead of the best result.

Make algorithm to match two strings

Print also:
Times there are no matches.
Specify the number of matches used to calculate the "cooked" stats.

There is a branch called optimization_try, which is an attempt to speed up the match algorithm.
Right now every genome is matched with every primer pair.
With the approach of this branch, every genome is split into chunks of n size and every chunk is matched with every primer in order to take advantage of memory locality. The matching on the forward primer is implemented (I think), but not on the reverse.

Use the program from terminal too

On forward start counting from the end.
Start always on position 1, not 0.

Matching parameters require enter to be updated while simulate's don't.

Use local versions of the libraries needed instead of linking them in order to make the program future proof.

No cases where there are more than one matches with same score. Low probability or bug?

Right now primer pair file requires an ID per primer pair, but the program uses a list, so the id makes no sense. Even worse, it can easily make the program crash. For example, if there are only 2 pairs, but with IDs 2 and 132 respectively, the program will try to get primer_pairs[131], which obviously will make the program crash!

Edit: It seems that only happens on the load_template function, my bad.
Edit2: And on GUI.

This bug has been corrected in the mload_files.py from master branch, but not in optimization_try's, which is likely to replace master's.

Check if dataframe.loc scales linearly as matrix size grows

When detecting missmatches pos and log Ns are detected as missmatches

Do not check for the genome file extension. Instead try to open every genome file as a fasta file.

Print a table containing genome and primer for each case where there have been no matches.

A "PrimerAlignment" instance is created even when no matches are found

I removed the header option when saving and now all headers are saved, ignoring the output info parameters