PASA是我们进行基因结构预测的时候常用的软件，其最早是开发用于拟南芥的基因结构注释，目前在动植物的注释中被广泛应用。

一、PASA使用准备文件：

1. genome.softmasked.fa # 软掩码之后的基因组
2. all_transcripts.fasta  # 有参、无参的转录组预测转录本；以及est.fa序列
3. all_transcripts.fasta.clean # 2的序列进行seqclean过滤之后的结果
4. FL_accs.txt # 全长转录本序列的ID名称
5. tdn.accs # denovo 组装的ID（无参转录本预测的序列）
6.  SQLite 或者是mysql数据库
7. alignAssembly.config # PASA运行的

二、PASA运行命令：

Launch_PASA_pipeline.pl  -c  -C -r -R -g  -T  -t   -u    -f  --TDN --CPU 24 --ALIGNERS blat,minimap2  --TRANSDECODER --MAX_INTRON_LENGTH 1000000 
# 注意，数据库的完整路径是写在了alignAssembly.config 这个配置文件里面

三、过程

1. create_sqlite_cdnaassembly_db.dbi -c alignAssembly.config -S 'PASApipeline/PASApipeline-v2.5.2/schema/cdna_alignment_sqliteschema' -r
samtools faidx all_transcripts.fasta.clean
2. PASApipeline/PASApipeline-v2.5.2/scripts/upload_transcript_data.dbi -M 'pasa.sqlite' -t all_transcripts.fasta.clean -T tdn.accs -f FL_accs.txt 
3. PASApipeline/PASApipeline-v2.5.2/scripts/run_spliced_aligners.pl --aligners blat,minimap2 --genome genome.softmasked.fa --transcripts all_transcripts.fasta.clean -I 1000000 -N 1 --CPU 24
4. PASApipeline/PASApipeline-v2.5.2/scripts/import_spliced_alignments.dbi -M 'pasa.sqlite'  -A blat -g blat.spliced_alignments.gff3
5. PASApipeline/PASApipeline-v2.5.2/scripts/import_spliced_alignments.dbi -M 'pasa.sqlite'  -A minimap2 -g minimap2.spliced_alignments.gff3
6. PASApipeline/PASApipeline-v2.5.2/pasa-plugins/transdecoder/TransDecoder.LongOrfs -t all_transcripts.fasta.clean  
7. PASApipeline/PASApipeline-v2.5.2/pasa-plugins/transdecoder/TransDecoder.Predict -t all_transcripts.fasta.clean 
8. PASApipeline/PASApipeline-v2.5.2/scripts/extract_FL_transdecoder_entries.pl all_transcripts.fasta.clean.transdecoder.gff3 > all_transcripts.fasta.clean.transdecoder.gff3.fl_accs
9. PASApipeline/PASApipeline-v2.5.2/scripts/update_fli_status.dbi -M 'pasa.sqlite' -f all_transcripts.fasta.clean.transdecoder.gff3.fl_accs
10. PASApipeline/PASApipeline-v2.5.2/scripts/validate_alignments_in_db.dbi -M 'pasa.sqlite' -g genome.softmasked.fa -t all_transcripts.fasta.clean --MAX_INTRON_LENGTH 1000000 --CPU 24   --MIN_PERCENT_ALIGNED 75  --NUM_BP_PERFECT_SPLICE_BOUNDARY 0  --MIN_AVG_PER_ID 95  > alignment.validations.output
11. PASApipeline/PASApipeline-v2.5.2/scripts/update_alignment_status.dbi -M 'pasa.sqlite' < alignment.validations.output  > pasa_run.log.dir/alignment.validation_loading.output
12. PASApipeline/PASApipeline-v2.5.2/scripts/PASA_transcripts_and_assemblies_to_GFF3.dbi -M pasa.sqlite' -v -A -P blat > pasa.sqlite.valid_blat_alignments.gff3
13. PASApipeline/PASApipeline-v2.5.2/scripts/PASA_transcripts_and_assemblies_to_GFF3.dbi -M 'pasa.sqlite' -v -A -P blat -B  > pasa.sqlite.valid_blat_alignments.bed
14. PASApipeline/PASApipeline-v2.5.2/scripts/PASA_transcripts_and_assemblies_to_GFF3.dbi -M 'pasa.sqlite' -v -A -P blat -T  > pasa.sqlite.valid_blat_alignments.gtf
15. PASApipeline/PASApipeline-v2.5.2/scripts/PASA_transcripts_and_assemblies_to_GFF3.dbi -M 'pasa.sqlite' -f -A -P blat > pasa.sqlite.failed_blat_alignments.gff3
16. /PASApipeline/PASApipeline-v2.5.2/scripts/PASA_transcripts_and_assemblies_to_GFF3.dbi -M 'pasa.sqlite' -f -A -P blat -B  > pasa.sqlite.failed_blat_alignments.bed
17. PASApipeline/PASApipeline-v2.5.2/scripts/PASA_transcripts_and_assemblies_to_GFF3.dbi -M 'pasa.sqlite' -f -A -P blat -T  > pasa.sqlite.failed_blat_alignments.gtf
18. PASApipeline/PASApipeline-v2.5.2/scripts/PASA_transcripts_and_assemblies_to_GFF3.dbi -M 'pasa.sqlite' -v -A -P minimap2 > pasa.sqlite.valid_minimap2_alignments.gff3
19. PASApipeline/PASApipeline-v2.5.2/scripts/PASA_transcripts_and_assemblies_to_GFF3.dbi -M 'pasa.sqlite' -v -A -P minimap2 -B  > pasa.sqlite.valid_minimap2_alignments.bed
20. PASApipeline/PASApipeline-v2.5.2/scripts/PASA_transcripts_and_assemblies_to_GFF3.dbi -M 'pasa.sqlite' -f -A -P minimap2 > pasa.sqlite.failed_minimap2_alignments.gff3
21. PASApipeline/PASApipeline-v2.5.2/scripts/PASA_transcripts_and_assemblies_to_GFF3.dbi -M 'pasa.sqlite' -f -A -P minimap2 -B  > pasa.sqlite.failed_minimap2_alignments.bed
22. PASApipeline/PASApipeline-v2.5.2/scripts/PASA_transcripts_and_assemblies_to_GFF3.dbi -M 'pasa.sqlite' -f -A -P minimap2 -T  > pasa.sqlite.failed_minimap2_alignments.gtf
23. PASApipeline/PASApipeline-v2.5.2/scripts/polyA_site_transcript_mapper.dbi -M 'pasa.sqlite' -c all_transcripts.fasta.cln -g genome.softmasked.fa -t all_transcripts.fasta > pasa_run.log.dir/polyAsite_analysis.out
24. PASApipeline/PASApipeline-v2.5.2/scripts/polyA_site_summarizer.dbi -M 'pasa.sqlite' -g genome.softmasked.fa  > pasa.sqlite.polyAsites.fasta
25. PASApipeline/PASApipeline-v2.5.2/scripts/reassign_clusters_via_valid_align_coords.dbi -M 'pasa.sqlite'  > pasa_run.log.dir/cluster_reassignment_by_valid_alignment_coords.default.out
26. PASApipeline/PASApipeline-v2.5.2/scripts/assemble_clusters.dbi -G genome.softmasked.fa  -M 'pasa.sqlite'  -T 24  > pasa.sqlite.pasa_alignment_assembly_building.ascii_illustrations.out
27. PASApipeline/PASApipeline-v2.5.2/scripts/polyA_site_summarizer.dbi -M 'pasa.sqlite' -g genome.softmasked.fa  > pasa.sqlite.polyAsites.fasta
28. PASApipeline/PASApipeline-v2.5.2/scripts/reassign_clusters_via_valid_align_coords.dbi -M 'pasa.sqlite'  > pasa_run.log.dir/cluster_reassignment_by_valid_alignment_coords.default.out
29. PASApipeline/PASApipeline-v2.5.2/scripts/assemble_clusters.dbi -G genome.softmasked.fa  -M 'pasa.sqlite'  -T 24  > pasa.sqlite.pasa_alignment_assembly_building.ascii_illustrations.out
30. PASApipeline/PASApipeline-v2.5.2/scripts/assembly_db_loader.dbi -M 'pasa.sqlite' > pasa_run.log.dir/alignment_assembly_loading.out
31. PASApipeline/PASApipeline-v2.5.2/scripts/subcluster_builder.dbi -G genome.softmasked.fa -M 'pasa.sqlite'   -m 50 > pasa_run.log.dir/alignment_assembly_subclustering.out
32. PASApipeline/PASApipeline-v2.5.2/scripts/populate_mysql_assembly_alignment_field.dbi -M 'pasa.sqlite' -G genome.softmasked.fa
33. PASApipeline/PASApipeline-v2.5.2/scripts/populate_mysql_assembly_sequence_field.dbi -M 'pasa.sqlite' -G genome.softmasked.fa
34. PASApipeline/PASApipeline-v2.5.2/scripts/subcluster_loader.dbi -M 'pasa.sqlite'  < pasa_run.log.dir/alignment_assembly_subclustering.out 
35. PASApipeline/PASApipeline-v2.5.2/scripts/alignment_assembly_to_gene_models.dbi -M 'pasa.sqlite' -G genome.softmasked.fa
36. PASApipeline/PASApipeline-v2.5.2/scripts/PASA_transcripts_and_assemblies_to_GFF3.dbi -M 'pasa.sqlite' -a  > pasa.sqlite.pasa_assemblies.gff3
37. PASApipeline/PASApipeline-v2.5.2/scripts/PASA_transcripts_and_assemblies_to_GFF3.dbi -M 'pasa.sqlite' -a -B  > pasa.sqlite.pasa_assemblies.bed
38. PASApipeline/PASApipeline-v2.5.2/scripts/PASA_transcripts_and_assemblies_to_GFF3.dbi -M 'pasa.sqlite' -a -T  > pasa.sqlite.pasa_assemblies.gtf
39. PASApipeline/PASApipeline-v2.5.2/scripts/describe_alignment_assemblies_cgi_convert.dbi -M 'pasa.sqlite'  > pasa.sqlite.pasa_assemblies_described.txt

在跑完PASA之后，关键结果就是pasa.sqlite.pasa_assemblies.gff3和pasa.sqlite.assemblies.fasta

四、出现的相关报错

1. 前一步的validate_alignments_in_db.dbi 没跑出来导致没有 alignment.validations.output这个文件

sh: alignment.validations.output: No such file or directory
Error, cmd: update_alignment_status.dbi -M 'pasa.sqlite'  < alignment.validations.output  > pasa_run.log.dir/alignment.validation_loading.output died with ret 256 No such file or directory at PASApipeline/PASApipeline-v2.5.2/PerlLib/Pipeliner.pm line 187

是因为配置文件里面写参数的时候后面加了#开头的注释，导致validate_alignments_in_db.dbi 命令里面存在注释，导致无法正常运行，删除注释之后运行无问题

2. 如果你的转录本是多个样本预测的合并结果（注意，这里是多个结果合并，而不是合并预测的结果），里面可能存在转录本ID一样的情况，此时也会报错。