Thanks to Haoran who reported the issue with fixed transcriptome read length generated by NanoSim for single thread #155
The reason for this issue was that the 2-dimensional Kernel Density Estimates (KDE) were generated all at once before selecting each reference transcript. Therefore, with a single thread, it would result in generating similar read lengths. Thanks to Haoran's suggestion, I adjusted the code so that it generate the 2D KDE everytime we select a reference transcript to simulate reads from, ensuring that there will be no similar read length given a specific reference transcript.
Please note that I did some analysis on different approaches to solve this issue and I believe the one suggested by Haoran (KDE generation for each reference transcript) is simple and efficient enough.
Thanks to Haoran who reported the issue with fixed transcriptome read length generated by NanoSim for single thread #155
The reason for this issue was that the 2-dimensional Kernel Density Estimates (KDE) were generated all at once before selecting each reference transcript. Therefore, with a single thread, it would result in generating similar read lengths. Thanks to Haoran's suggestion, I adjusted the code so that it generate the 2D KDE everytime we select a reference transcript to simulate reads from, ensuring that there will be no similar read length given a specific reference transcript.
Please note that I did some analysis on different approaches to solve this issue and I believe the one suggested by Haoran (KDE generation for each reference transcript) is simple and efficient enough.