Need code for my problem

Myseq*.tsv file is a output file. I need file in this format, this is matrix form DUplex_subset is a input file. For example: >hsa-miR-576-3p ENST00000379407 Forward: Score: 149.000000 Q:2 to 18 R:2388 to 2409 Align Len (16) (75.00%) (87.50%) Energy:-9.900000 (mRNA) Ref: 5' aagcaTAATTCTTACATATCTg 3' :|||| || ||:|||| (miRNA) Query: 3' cuaagGUUAAAAAGGUGUAGAa 5' Here Query is miRNA and Ref are mRNA. Extract only capital letter alignment is seed region and the lowercase alphabet is non-seed region. I need to extract each alignment of seed only and put it in matrix form and give number to each as like in output file Output: Note: extract only which is in bold letters in each column: miRNA mRNA R Pos1 Pos2 Pos3 Pos4 Pos5 Pos6 Pos7 hsa-miR-576-3p ENST00000379407 2388 to 2409 TA GC TA AU TU AG AG Give the number to each alignment pairing: Watson and Crick(GC, AU)-- 4 mismatch--- -1 GU ------ 1 Other mismatches ------ 0 Gap -------- -2 i sent my practice code also