[March 2020] The Aspergillus flavus NRRL 3357 genome was not sequenced and assembled at the Joint Genome Institute, but rather by Jeff Skerker using a combination of long-read and short-read datasets (Pacbio, Oxford Nanopore, and Illumina).  Eight chromosomes were assembled using a hybrid assembly method and the CANU assembler (v1.7.1).  The assembly was polished using Pacbio data using pbalign (v0.3.1), blasr (v5.3) and the Arrow (v2.2.2) algorithm. Final error correction performed using Pilon and Illumina data.  Eight chromosomes were assembled, 7 out of 8 are complete telomere-to-telomere assemblies.

This assembly was then annotated using the JGI annotation pipeline, with several modifications. To preserve original gene names as much as possible, previously produced models available on fungiDB were mapped forward onto the new assembly, which constitute 43.29% of the Filtered Model (FM) set. Additionally, to improve capture of UTRs using RNAseq data available at NCBI, we applied our est extension procedure to mapped fungiDB models, producing a new track, estExt_Aspflav1_ExternalModels (20.79% of FM). Our standard filtering parameters were also adjusted to allow capture of more models with transcriptomic support, as well as prioritize mapped forward models (and their est extended versions) for inclusion in the FM set.