作用：对 GETA 基因模型做跨工具过滤——合并 RNA-seq 文库，用 BLASTP 与 Pfam 筛查预测蛋白，并估计转录丰度，据此保留有表达支持的基因模型。
入口：cat.sh、run_BLASTP.sh、run_PfamScan.sh、run_RSEM.sh。
输入：.../data/Gifu/RNA/clean/1 与 /2 中列出的各文库 clean reads（由 cat.sh 合并）；Gifu T2T 预测蛋白 Gifu_T2T.protein.fasta（来自步骤 05）；BLASTP 命令列表 cmd.list；RSEM 所需的 reference.fasta、reference.fasta.fai、gene.map。
输出：Gifu_RNA_all_R1.fastq.gz、Gifu_RNA_all_R2.fastq.gz；blast*.out.tmp → BLASTP.OUT.TMP；pfam.out；rsem_outdir/（含 bowtie.bam）、sorted.bam、gene.bed、transcriptome_chromSizes.txt、cov.bed。
运行：bash cat.sh → bash run_BLASTP.sh → bash run_PfamScan.sh → bash run_RSEM.sh（cat.sh 必须先跑，run_RSEM.sh 消费它生成的合并 fastq）。
工具：ParaFly -c cmd.list -CPU 80；pfam_scan.pl -fasta Gifu_T2T.protein.fasta -dir <PfamScan/data> -cpu 88 -outfile pfam.out；align_and_estimate_abundance.pl --est_method RSEM --aln_method bowtie --prep_reference --thread_count 64 加 bowtie；samtools sort -@ 64；bedtools coverage -a gene.bed -b sorted.bam -sorted -g transcriptome_chromSizes.txt；BLASTP 参数写在 cmd.list 里。
English
**Purpose**: Cross-tool filtering of the GETA gene models — pool the RNA-seq libraries, screen the predicted proteins by BLASTP and against Pfam, and estimate transcript abundance so that gene models are kept on expression support.
**Entry point**: `cat.sh`, `run_BLASTP.sh`, `run_PfamScan.sh`, `run_RSEM.sh`.
**Inputs**: the clean reads of the libraries listed in `.../data/Gifu/RNA/clean/1` and `/2` (pooled by `cat.sh`); Gifu T2T predicted proteins `Gifu_T2T.protein.fasta` (from step 05); `cmd.list`, the BLASTP command list; `reference.fasta`, `reference.fasta.fai` and `gene.map`, required by RSEM.
**Outputs**: `Gifu_RNA_all_R1.fastq.gz` and `Gifu_RNA_all_R2.fastq.gz`; `blast*.out.tmp` → `BLASTP.OUT.TMP`; `pfam.out`; `rsem_outdir/` (containing `bowtie.bam`), `sorted.bam`, `gene.bed`, `transcriptome_chromSizes.txt`, `cov.bed`.
**Run**: `bash cat.sh` → `bash run_BLASTP.sh` → `bash run_PfamScan.sh` → `bash run_RSEM.sh` (`cat.sh` must run first, since `run_RSEM.sh` consumes the pooled fastq files it writes).
**Tools**: ParaFly `-c cmd.list -CPU 80`; `pfam_scan.pl -fasta Gifu_T2T.protein.fasta -dir -cpu 88 -outfile pfam.out`; `align_and_estimate_abundance.pl --est_method RSEM --aln_method bowtie --prep_reference --thread_count 64` with bowtie; `samtools sort -@ 64`; `bedtools coverage -a gene.bed -b sorted.bam -sorted -g transcriptome_chromSizes.txt`; BLASTP parameters live in `cmd.list`.