SNAP prediction failed and overlaps exon of the .zff file

[augauga]

Thank you very much for developing funannotate! I encountered a problem when predicting genes. I used funannotate v1.8.17 installed by conda and my command line was like:

funannotate predict -i contigs.masked.fasta --transcript_evidence merged_EST.fna -o ./ --augustus_species saccharomyces --cpus 2 --species Candida_auris

The pipeline worked well and I got some results. But when I checked the log file, I found the SNAP predicted 0 genes and it was deleted in the EVM weights and json file.

logfile

[10/23/24 21:32:56]: gene175.t1 1 1 2 - errors(1): exon-1:overlaps_prev_exon
gene1052.t1 1 1 2 + errors(1): exon-2:overlaps_prev_exon
gene870.t1 1 1 2 + errors(1): exon-2:overlaps_prev_exon
gene214.t1 1 1 2 - errors(1): exon-1:overlaps_prev_exon
gene303.t1 1 1 2 - errors(1): exon-1:overlaps_prev_exon
gene963.t1 1 1 3 + errors(1): exon-3:overlaps_prev_exon
gene255.t1 1 1 2 + errors(1): exon-2:overlaps_prev_exon
gene483.t1 1 1 10 + errors(1): exon-10:overlaps_prev_exon
gene511.t1 1 1 2 + errors(1): exon-2:overlaps_prev_exon
gene1165.t1 1 1 3 + errors(1): exon-3:overlaps_prev_exon
......
[10/23/24 21:32:56]: fathom uni.ann uni.dna -export 1000 -plus
[10/23/24 21:32:56]: forge export.ann export.dna
[10/23/24 21:32:56]: perl /home/jerry/miniconda3/envs/funannotate/bin/hmm-assembler.pl snap-trained ./predict_misc/snaptrain
[10/23/24 21:32:56]: snap /mnt/data/bear/jerry/DR_project/1_fastq/Australia/SRR7657927/5_annotate/2_prediction/predict_misc/snap-trained.hmm /mnt/data/bear/jerry/DR_project/1_fastq/Australia/SRR7657927/5_annotate/2_prediction/predict_misc/genome.softmasked.fa
[10/23/24 21:33:35]: scoring....decoding.10.20.30.40.50.60.70.80.90.100 done
scoring....decoding.10.20.30.40.50.60.70.80.90.100 done
scoring....decoding.10.20.30.40.50.60.70.80.90.100 done
scoring....decoding.10.20.30.40.50.60.70.80.90.100 done
scoring....decoding.10.20.30.40.50.60.70.80.90.100 done
scoring....decoding.10.20.30.40.50.60.70.80.90.100 done
scoring....decoding.10.20.30.40.50.60.70.80.90.100 done
scoring....decoding.10.20.30.40.50.60.70.80.90.100 done
scoring....decoding.10.20.30.40.50.60.70.80.90.100 done
scoring....decoding.10.20.30.40.50.60.70.80.90.100 done
scoring....decoding.10.20.30.40.50.60.70.80.90.100 done
...
[10/23/24 21:33:35]: 0 predictions from SNAP
[10/23/24 21:33:35]: SNAP prediction failed, moving on without result
[10/23/24 21:33:35]: snap failed removing from training parameters

I also noticed this issue #386 and try to install SNAP manually but it still didn't work. After checking the input of SNAP, "snap.training.zff" and I am sure something wrong with the input files.

>scaffold_1
Einit	283	2340	gene255.t1
Eterm	2338	2340	gene255.t1
Einit	4575	2389	gene175.t1
Eterm	2391	2389	gene175.t1
Einit	18732	19758	gene963.t1
Exon	19819	21482	gene963.t1
Eterm	21480	21482	gene963.t1
Einit	47523	48835	gene483.t1
Exon	49966	51130	gene483.t1
Exon	51233	51298	gene483.t1
Exon	51341	51356	gene483.t1
Exon	51414	51649	gene483.t1
Exon	51873	51991	gene483.t1
Exon	52445	52455	gene483.t1
Exon	52851	53037	gene483.t1
Exon	53674	53692	gene483.t1
Eterm	53690	53692	gene483.t1
Einit	106697	107713	gene1052.t1
Eterm	107711	107713	gene1052.t1
Einit	111697	109787	gene303.t1
Eterm	109789	109787	gene303.t1
Einit	116365	117687	gene870.t1
Eterm	117685	117687	gene870.t1
Einit	131743	129929	gene214.t1
Eterm	129931	129929	gene214.t1
Einit	175298	176155	gene511.t1
Eterm	176153	176155	gene511.t1
......

And the "busco.final.gff3" file seems abnormal too:

scaffold_1	Augustus	gene	283	2340	.	+	.	ID=gene255;Name=EOG09261Q18;
scaffold_1	Augustus	mRNA	283	2340	.	+	.	ID=gene255.t1;Parent=gene255;product=hypothetical protein;
scaffold_1	Augustus	exon	283	2340	.	+	.	ID=gene255.t1.exon1;Parent=gene255.t1;
scaffold_1	Augustus	exon	2338	2340	.	+	.	ID=gene255.t1.exon2;Parent=gene255.t1;
scaffold_1	Augustus	CDS	283	2340	.	+	0	ID=gene255.t1.cds;Parent=gene255.t1;
scaffold_1	Augustus	CDS	2338	2340	.	+	0	ID=gene255.t1.cds;Parent=gene255.t1;
scaffold_1	Augustus	gene	2389	4575	.	-	.	ID=gene175;Name=EOG09261BPJ;
scaffold_1	Augustus	mRNA	2389	4575	.	-	.	ID=gene175.t1;Parent=gene175;product=hypothetical protein;
scaffold_1	Augustus	exon	2389	4575	.	-	.	ID=gene175.t1.exon1;Parent=gene175.t1;
scaffold_1	Augustus	exon	2389	2391	.	-	.	ID=gene175.t1.exon2;Parent=gene175.t1;
scaffold_1	Augustus	CDS	2389	4575	.	-	0	ID=gene175.t1.cds;Parent=gene175.t1;
scaffold_1	Augustus	CDS	2389	2391	.	-	0	ID=gene175.t1.cds;Parent=gene175.t1;
scaffold_1	Augustus	gene	18732	21482	.	+	.	ID=gene963;Name=EOG092628LW;
scaffold_1	Augustus	mRNA	18732	21482	.	+	.	ID=gene963.t1;Parent=gene963;product=hypothetical protein;
scaffold_1	Augustus	exon	18732	19758	.	+	.	ID=gene963.t1.exon1;Parent=gene963.t1;
scaffold_1	Augustus	exon	19819	21482	.	+	.	ID=gene963.t1.exon2;Parent=gene963.t1;
scaffold_1	Augustus	exon	21480	21482	.	+	.	ID=gene963.t1.exon3;Parent=gene963.t1;
scaffold_1	Augustus	CDS	18732	19758	.	+	0	ID=gene963.t1.cds;Parent=gene963.t1;
scaffold_1	Augustus	CDS	19819	21482	.	+	2	ID=gene963.t1.cds;Parent=gene963.t1;
scaffold_1	Augustus	CDS	21480	21482	.	+	0	ID=gene963.t1.cds;Parent=gene963.t1;
......

I have just listed the head of these files. And the problem is the gff3 file add an extra line for each gene as a exon, where it should be a stop codon. So that snap cannot recognize the genes and got empty "uni.ann". I think removing these lines manually will solve this problem. But I don't know how to fix it in the pipeline. So, do you have any idea about this problem and how can I make the snap work again?

Thanks very much. I am looking forward to your response.