Transcriptome
Eric Ross
Overview
Assembly and annotation of Mnemiopsis leidyi transcriptome.
Data Sources
| Time point | Year Collected | Replicate Name | left(1)/right(2) | Filename |
|---|---|---|---|---|
| 1hpf | 2012 | N7 | 1 | 1hpf_2012_N7_1.fastq.gz |
| 1hpf | 2012 | N7 | 2 | 1hpf_2012_N7_2.fastq.gz |
| 1hpf | 2013 | 21 | 1 | 1hpf_2013_21_1.fastq.gz |
| 1hpf | 2013 | 21 | 2 | 1hpf_2013_21_2.fastq.gz |
| 1hpf | 2013 | 41 | 1 | 1hpf_2013_41_1.fastq.gz |
| 1hpf | 2013 | 41 | 2 | 1hpf_2013_41_2.fastq.gz |
| 2hpf | 2012 | N6 | 1 | 2hpf_2012_N6_1.fastq.gz |
| 2hpf | 2012 | N6 | 2 | 2hpf_2012_N6_2.fastq.gz |
| 2hpf | 2013 | 22 | 1 | 2hpf_2013_22_1.fastq.gz |
| 2hpf | 2013 | 22 | 2 | 2hpf_2013_22_2.fastq.gz |
| 2hpf | 2013 | 32 | 1 | 2hpf_2013_32_1.fastq.gz |
| 2hpf | 2013 | 32 | 2 | 2hpf_2013_32_2.fastq.gz |
| 2hpf | 2013 | 42 | 1 | 2hpf_2013_42_1.fastq.gz |
| 2hpf | 2013 | 42 | 2 | 2hpf_2013_42_2.fastq.gz |
| 3hpf | 2012 | N5 | 1 | 3hpf_2012_N5_1.fastq.gz |
| 3hpf | 2012 | N5 | 2 | 3hpf_2012_N5_2.fastq.gz |
| 3hpf | 2013 | 23 | 1 | 3hpf_2013_23_1.fastq.gz |
| 3hpf | 2013 | 23 | 2 | 3hpf_2013_23_2.fastq.gz |
| 3hpf | 2013 | 33 | 1 | 3hpf_2013_33_1.fastq.gz |
| 3hpf | 2013 | 33 | 2 | 3hpf_2013_33_2.fastq.gz |
| 4hpf | 2012 | N4 | 1 | 4hpf_2012_N4_1.fastq.gz |
| 4hpf | 2012 | N4 | 2 | 4hpf_2012_N4_2.fastq.gz |
| 4hpf | 2013 | 24 | 1 | 4hpf_2013_24_1.fastq.gz |
| 4hpf | 2013 | 24 | 2 | 4hpf_2013_24_2.fastq.gz |
| 4hpf | 2013 | 34 | 1 | 4hpf_2013_34_1.fastq.gz |
| 4hpf | 2013 | 34 | 2 | 4hpf_2013_34_2.fastq.gz |
| 5hpf | 2012 | 14 | 1 | 5hpf_2012_14_1.fastq.gz |
| 5hpf | 2012 | 14 | 2 | 5hpf_2012_14_2.fastq.gz |
| 5hpf | 2012 | N3 | 1 | 5hpf_2012_N3_1.fastq.gz |
| 5hpf | 2012 | N3 | 2 | 5hpf_2012_N3_2.fastq.gz |
| 5hpf | 2013 | 25 | 1 | 5hpf_2013_25_1.fastq.gz |
| 5hpf | 2013 | 25 | 2 | 5hpf_2013_25_2.fastq.gz |
| 6hpf | 2012 | 23 | 1 | 6hpf_2012_23_1.fastq.gz |
| 6hpf | 2012 | 23 | 2 | 6hpf_2012_23_2.fastq.gz |
| 6hpf | 2012 | N2 | 1 | 6hpf_2012_N2_1.fastq.gz |
| 6hpf | 2012 | N2 | 2 | 6hpf_2012_N2_2.fastq.gz |
| 6hpf | 2013 | 26 | 1 | 6hpf_2013_26_1.fastq.gz |
| 6hpf | 2013 | 26 | 2 | 6hpf_2013_26_2.fastq.gz |
| 6hpf | 2013 | 46 | 1 | 6hpf_2013_46_1.fastq.gz |
| 6hpf | 2013 | 46 | 2 | 6hpf_2013_46_2.fastq.gz |
| 7hpf | 2012 | N1 | 1 | 7hpf_2012_N1_1.fastq.gz |
| 7hpf | 2012 | N1 | 2 | 7hpf_2012_N1_2.fastq.gz |
| 7hpf | 2013 | 17 | 1 | 7hpf_2013_17_1.fastq.gz |
| 7hpf | 2013 | 17 | 2 | 7hpf_2013_17_2.fastq.gz |
| 7hpf | 2013 | 27 | 1 | 7hpf_2013_27_1.fastq.gz |
| 7hpf | 2013 | 27 | 2 | 7hpf_2013_27_2.fastq.gz |
| 7hpf | 2013 | 47 | 1 | 7hpf_2013_47_1.fastq.gz |
| 7hpf | 2013 | 47 | 2 | 7hpf_2013_47_2.fastq.gz |
| 8hpf | 2012 | 18 | 1 | 8hpf_2012_18_1.fastq.gz |
| 8hpf | 2012 | 18 | 2 | 8hpf_2012_18_2.fastq.gz |
| 8hpf | 2012 | N8B | 1 | 8hpf_2012_N8B_1.fastq.gz |
| 8hpf | 2012 | N8B | 2 | 8hpf_2012_N8B_2.fastq.gz |
| 8hpf | 2013 | 18 | 1 | 8hpf_2013_18_1.fastq.gz |
| 8hpf | 2013 | 18 | 2 | 8hpf_2013_18_2.fastq.gz |
| 8hpf | 2013 | 28 | 1 | 8hpf_2013_28_1.fastq.gz |
| 8hpf | 2013 | 28 | 2 | 8hpf_2013_28_2.fastq.gz |
| 8hpf | 2013 | 48 | 1 | 8hpf_2013_48_1.fastq.gz |
| 8hpf | 2013 | 48 | 2 | 8hpf_2013_48_2.fastq.gz |
| 9hpf | 2012 | 37 | 1 | 9hpf_2012_37_1.fastq.gz |
| 9hpf | 2012 | 37 | 2 | 9hpf_2012_37_2.fastq.gz |
| 9hpf | 2013 | 49 | 1 | 9hpf_2013_49_1.fastq.gz |
| 9hpf | 2013 | 49 | 2 | 9hpf_2013_49_2.fastq.gz |
Assembly
Left and right reads were separately concatenated and then assembled with Trinity version trinityrnaseq_r20140717 (Grabherr et al. 2011).
Trinity \
--normalize_reads \
--max_memory 500G \
--CPU 16 \
--SS_lib_type FR \
--left \
lefts.fastq \
--right \
rights.fastq
Initial Assembly Statistics
| Number of Sequences: | 233,327 |
| Total Length: | 160,677,750 |
| Average Length: | 688 |
| Longest Sequence: | 29,348 |
| Shortest Sequence: | 201 |
| %GC: | 40% |
| N50: | 1,152 |
| N90: | 266 |
Sequence Cleaning
Adapter and contaminate was removed from assembly using SeqClean (Cleaner 2017). Additionally, sequences shorter than 300 bp were discarded.
seqclean \
Mnemiopsis_leidyi_trinity_20140802_unfiltered.nt \
-v UniVec \
-l 300 \
-c 16 \
-s Hs_GRCh38.nt,ecoli.nt
Output:
**************************************************
Sequences analyzed: 233327
-----------------------------------
valid: 141346 (24371 trimmed)
trashed: 91981
**************************************************
----= Trashing summary =------
by 'short': 90500
by 'Hs_GRCh38.nt': 11
by 'shortq': 140
by 'ecoli.nt': 17
by 'UniVec': 1313
------------------------------
Cleaned Assembly Statistics
| Number of Sequences: | 141,346 |
| Total Length: | 137,906,212 |
| Average Length: | 975 |
| Longest Sequence: | 29,348 |
| Shortest Sequence: | 300 |
| %GC: | 41% |
| N50: | 1,486 |
| N90: | 405 |
Genome Filtering
In order to produce a more accurate transcriptome set, we filtered out sequences not present in the available M. leidyi genome. We used the very permissive requirement that any transcript aligning with at least 90% identity over any length of genomic sequence ought to be retained.
Genome Filtered Transcript Statistics
| Number of Sequences: | 44,438 |
| Total Length: | 62,917,788 |
| Average Length: | 1,415 |
| Longest Sequence: | 29,348 |
| Shortest Sequence: | 300 |
| %GC: | 41% |
| %N: | 0% |
| N50: | 2,168 |
| N90: | 601 |
Additional Sequence Cleaning
NCBI detected additional contaminate in our sequences, primarily vector/adapter. We sequentially ran SeqClean to remove concatenated adapter.
Cleaned Sequence Statistics
| Number of Sequences: | 44,388 |
| Total Length: | 62,859,653 |
| Average Length: | 1,416 |
| Longest Sequence: | 29,348 |
| Shortest Sequence: | 300 |
| %GC: | 41% |
| %N: | 0% |
| N50: | 2,167 |
| N90: | 601 |
Transcript collapse
For RNAseq we wished to create a non-redundant set of transcripts by collapsing isoforms using CD-HIT (Li, Jaroszewski, and Godzik 2001).
cd-hit-est \
-i in_genome_clean_r6.nt \
-o cdhit.nt \
-r 0 \
-c .95 \
-G 0 \
-aS .5 \
-M 0 \
-T 0
Non-redundant Sequence Statistics
| Number of Sequences: | 31,067 |
| Total Length: | 47,344,147 |
| Average Length: | 1,523 |
| Longest Sequence: | 29,348 |
| Shortest Sequence: | 300 |
| %GC: | 41% |
| %N: | 0% |
| N50: | 2,200 |
| N90: | 692 |
Sequence Renaming
Sequences were renamed with a “Mle” prefix and zero padded numbers. Names are in the format of ‘Mle_000000’.
Annotation
Swissprot / Uniprot
Best BLASTX hits to Swissprot with a maximum evalue of 0.001 were used to annotate genes (Camacho et al. 2009).
blastx \
-query Mleidyi_20160525.nt \
-db swissprot \
-max_target_seqs 1 \
-max_hsps 1 \
-evalue .001 \
-outfmt 6 \
-out ml_v_sp.blastx \
-num_threads 32
Pfam
Sequences were translated with Transdecoder and then domains were assigned to transcripts using hmmscan against the Pfam-A database (Eddy 2011).Transdecoder.LongOrfs \
-S \
-m 60 \
-t Mleidyi_20160525.aa.fa
All ORFs of 60aa or greater were retained.
hmmscan \
-o Mleidyi_20160525.aa.fa.pfam.out \
--domtblout Mleidyi_20160525.aa.fa.pfam.table \
-E .01 \
--domE .01 \
--cpu 16 \
../db/Pfam-A.hmm \
awk '{print $4 "\t" $2}' Mleidyi_20160525.aa.fa.pfam.table | perl -p -e 's/(Mle_\d+)\|.*(\t.+)/$1$2/' | sort -u > mleidyi2pfam.txt
Software
| Program | Version |
|---|---|
| Trinity | trinityrnaseq_r20140717 |
| Seqclean | Downloaded Feb 22, 2011 |
| BLAST+ | 2.3.0 |
| Transdecoder | 2.0.1 |
| cd-hit | CD-HIT version 4.6 |
| HMMeR | 3.1b2 |
Accession Numbers
BioProject: PRJNA344880
TSA: GFAT00000000
References
Camacho, Christiam, George Coulouris, Vahram Avagyan, Ning Ma, Jason Papadopoulos, Kevin Bealer, and Thomas L Madden. 2009. “BLAST+: Architecture and Applications.” BMC Bioinformatics 10 (1). Springer Nature: 421. doi:10.1186/1471-2105-10-421.
Cleaner. 2017. “Sequence Cleaner.” SourceForge. https://sourceforge.net/projects/seqclean/.
Eddy, Sean R. 2011. “Accelerated Profile HMM Searches.” Edited by William R. Pearson. PLoS Computational Biology 7 (10). Public Library of Science (PLoS): e1002195. doi:10.1371/journal.pcbi.1002195.
Grabherr, Manfred G, Brian J Haas, Moran Yassour, Joshua Z Levin, Dawn A Thompson, Ido Amit, Xian Adiconis, et al. 2011. “Full-Length Transcriptome Assembly from RNA-Seq Data Without a Reference Genome.” Nature Biotechnology 29 (7). Springer Nature: 644–52. doi:10.1038/nbt.1883.
Li, W., L. Jaroszewski, and A. Godzik. 2001. “Clustering of Highly Homologous Sequences to Reduce the Size of Large Protein Databases.” Bioinformatics 17 (3). Oxford University Press (OUP): 282–83. doi:10.1093/bioinformatics/17.3.282.
