The discrimination of exons from introns in the DNA sequence of a eukaryotic gene is important to understand the functionality of protein formation inside a living organism. Several signal processing techniques involving transforms and filtering have been used to identify the exon regions by exploring the periodicity‐3 property. Fast processing of massive DNA sequences is desirable to detect the disease‐causing mechanism, which is helpful to prepare individual‐centric drugs. In this manuscript, a hardware implementation is carried out for the direct form‐II structure of the infinite impulse response anti‐notch filter to achieve the fast processing of the DNA sequence. Implementation result on Zynq‐series (Zybo board) Field Programmable Gate Array (FPGA) reveals that the proposed implementation is capable to identify the exon regions of five benchmark eukaryotic genes. The FPGA implementation has achieved a maximum clock frequency of 34.629 MHz, which is further improved to 54.41 MHz using the retiming concept. Compared with MATLAB 2014a simulation, the proposed FPGA implementation has achieved similar accuracy with 39 to 43 times faster computing time for the five benchmark data sets. Further, an Application Specific Integrated Circuit (ASIC) implementation is carried out in the CADENCE Register Transfer Level (RTL) compiler tool with GPDK 90‐nm technology, due to which the hardware anti‐notch filter is 120 to 133 times faster compared with its MATLAB counterpart while maintaining the comparable accuracy.

中文翻译：

---

无限脉冲响应抗陷波滤波器在真核基因外显子区域识别中的硬件实现

真核基因DNA序列中内含子与外显子的区别对于理解生物体内蛋白质形成的功能非常重要。通过探索periodicity-3属性，已使用几种涉及变换和滤波的信号处理技术来识别外显子区域。快速检测大量DNA序列对于检测致病机制是理想的，这有助于制备以个体为中心的药物。在此手稿中，对无限冲激响应抗陷波滤波器的直接形式II结构进行了硬件实现，以实现DNA序列的快速处理。在Zynq系列（Zybo板）现场可编程门阵列（FPGA）上的实施结果表明，该实施方案能够识别五个基准真核基因的外显子区域。FPGA实现已实现了34.629 MHz的最大时钟频率，使用重定时概念可以将其进一步提高到54.41 MHz。与MATLAB 2014a仿真相比，拟议的FPGA实现实现了相似的精度，对五个基准数据集的计算时间加快了39到43倍。此外，在具有GPDK 90-nm技术的CADENCE寄存器传输级别（RTL）编译器工具中执行了专用集成电路（ASIC）实施，由于该硬件，其抗陷波滤波器的速度是其抗陷波滤波器的120到133倍。同时保持与MATLAB相当的准确性。