bioinformatics - fastq 文件中的簇位置单位

Question

我有一个 fastq 文件，其 header 包含与 wikipedia example 格式相同的位置。对于fastq:

@HWUSI-EAS100R:6:73:941:1973#0/1

我知道 x=941 和 y=1973，但是单位是什么？微米？一些照明距离？

如果在 x=931 和 y=1973 处存在另一个簇，它们之间的距离是否为 10 微米？

谢谢

编辑: 联系了 Illumina 支持人员，他们说我需要的信息(例如平铺查看区域的尺寸或显微镜的变焦)“被认为是专有的”，这令人沮丧。

Answer 1

作为短期解决方案，我使用 here 提供的最佳簇密度来估计 1 像素(以微米为单位)的大小。

我绘制了给定图 block 中簇的 (x,y) 位置，以获得如下图:

步骤:

1. 根据存在的簇数量估计观察窗口的面积
2. 从 mm^2 转换为 um^2
3. 将 um^2 转换为窗口半径(以 um 为单位)
4. 计算窗口沿 x 维度的半径(以像素为单位)
5. 将两者相除即可得到每微米的像素数

代码:

print 'For {} clusters, assuming optimal is {}'.format(num_clusters, num_optimal)
for perc in perc_optimal:
    mm_sqr = num_clusters/(perc*num_optimal)
    um_sqr = mm_sqr*1e6
    um_radius = np.sqrt(um_sqr/np.pi)

    px_radius = (pos['x'].max()-pos['x'].min())/2
    px_to_um = px_radius/um_radius

    out = ('At {}% optimal, tile r is {} um, pixel r is {} px,'
           'so {} px is 1 um'.format(perc*100,um_radius,px_radius,px_to_um))

    print out

输出:

For 531143 clusters, assuming optimal is 900000
At 10.0% optimal, tile r is 1370.59625256 um, pixel r is 13997 px,so 10.2123436963 px is 1 um
At 20.0% optimal, tile r is 969.157904453 um, pixel r is 13997 px,so 14.4424349589 px is 1 um
At 30.0% optimal, tile r is 791.314115365 um, pixel r is 13997 px,so 17.6882981464 px is 1 um
At 40.0% optimal, tile r is 685.298126279 um, pixel r is 13997 px,so 20.4246873926 px is 1 um
At 50.0% optimal, tile r is 612.949278085 um, pixel r is 13997 px,so 22.8354947145 px is 1 um
At 60.0% optimal, tile r is 559.543577023 um, pixel r is 13997 px,so 25.0150311339 px is 1 um
At 70.0% optimal, tile r is 518.036690306 um, pixel r is 13997 px,so 27.0193217236 px is 1 um
At 80.0% optimal, tile r is 484.578952226 um, pixel r is 13997 px,so 28.8848699179 px is 1 um
At 90.0% optimal, tile r is 456.865417519 um, pixel r is 13997 px,so 30.6370310889 px is 1 um
At 100.0% optimal, tile r is 433.420591057 um, pixel r is 13997 px,so 32.2942663288 px is 1 um

这一切都非常粗糙，我希望有一个不同的答案，但这也许对某人有帮助