Many SNPs with low PIP in credible set

[PhoebeGuo97]

Hello, I ran susie_rss for a region of chromosome 17 with ~5000 SNPs and got 1631 SNPs in the output. The highest PIP is ~0.0027. I'm wondering if I should change some parameters to reduce the SNPs in the CS. Thanks!

rss$sets is as below: $cs $cs$L1 [1] 1069 1092 1093 1121 1122 1123 1124 1126 1127 1128 1129 1130 1131 1134 1135 1138 1139 1143 1149 1151 1152 [22] 1153 1155 1156 1157 1159 1160 1162 1164 1168 1171 1172 1176 1178 1180 1181 1184 1185 1186 1187 1188 1189 [43] 1190 1191 1194 1195 1197 1201 1202 1203 1204 1206 1209 1211 1213 1215 1216 1217 1218 1221 1222 1223 1226 [64] 1228 1232 1233 1234 1242 1243 1244 1246 1249 1258 1260 1261 1262 1263 1265 1266 1268 1270 1271 1275 1276 [85] 1277 1279 1283 1286 1287 1288 1290 1291 1293 1296 1298 1300 1301 1302 1303 1304 1305 1306 1307 1308 1312 [106] 1313 1315 1316 1317 1319 1321 1325 1327 1329 1330 1332 1333 1334 1335 1336 1337 1338 1339 1340 1342 1344 [127] 1345 1347 1349 1350 1351 1353 1354 1355 1356 1357 1360 1361 1365 1366 1367 1368 1369 1370 1371 1372 1373 [148] 1374 1375 1378 1379 1381 1382 1383 1384 1385 1386 1388 1389 1391 1393 1396 1397 1399 1400 1401 1404 1405 [169] 1407 1408 1410 1411 1412 1414 1416 1417 1419 1421 1422 1424 1428 1430 1431 1432 1434 1435 1436 1437 1438 [190] 1440 1442 1443 1444 1445 1447 1448 1450 1451 1452 1453 1454 1455 1456 1457 1458 1459 1461 1462 1463 1464 [211] 1465 1466 1468 1469 1474 1476 1479 1480 1481 1482 1483 1484 1486 1488 1489 1490 1491 1492 1495 1497 1499 [232] 1500 1502 1503 1505 1507 1509 1510 1511 1512 1513 1520 1638 1640 1641 1642 1643 1644 1645 1646 1647 1648 [253] 1649 1650 1651 1654 1655 1656 1657 1658 1661 1662 1665 1666 1667 1672 1674 1675 1676 1677 1683 1684 1686 [274] 1687 1688 1689 1690 1692 1693 1694 1695 1696 1698 1699 1701 1702 1705 1707 1710 1711 1717 1718 1719 1721 [295] 1723 1724 1725 1727 1728 1729 1730 1731 1732 1734 1735 1736 1737 1738 1741 1744 1745 1746 1747 1748 1753 [316] 1755 1758 1760 1761 1762 1763 1764 1765 1766 1767 1769 1771 1772 1773 1774 1776 1778 1779 1780 1781 1783 [337] 1784 1786 1787 1789 1794 1796 1799 1801 1802 1805 1806 1807 1809 1810 1811 1812 1814 1816 1817 1819 1820 [358] 1822 1823 1824 1825 1826 1827 1828 1829 1832 1834 1835 1836 1837 1839 1841 1842 1845 1848 1856 1857 1858 [379] 1861 1864 1869 1872 1873 1875 1876 1877 1882 1884 1885 1886 1887 1889 1892 1896 1898 1900 1901 1905 1906 [400] 1908 1910 1911 1912 1913 1916 1917 1918 1922 1924 1925 1927 1928 1932 1933 1935 1937 1939 1940 1941 1942 [421] 1945 1946 1948 1951 1953 1957 1959 1960 1961 1964 1967 1968 1969 1970 1971 1980 1982 1984 1985 1987 1988 [442] 1989 1991 1992 1994 1996 1997 1998 1999 2001 2004 2007 2008 2010 2011 2012 2014 2017 2020 2025 2026 2027 [463] 2029 2030 2031 2032 2033 2034 2036 2037 2038 2040 2044 2055 2056 2057 2058 2059 2066 2069 2070 2072 2074 [484] 2077 2078 2082 2138 2151 2161 2168 2186 2187 2188 2190 2192 2194 2195 2197 2199 2200 2206 2207 2208 2209 [505] 2211 2212 2213 2214 2215 2217 2219 2221 2222 2223 2226 2227 2228 2231 2236 2239 2249 2250 2251 2253 2256 [526] 2257 2258 2261 2262 2263 2267 2268 2269 2273 2274 2275 2280 2281 2282 2284 2286 2288 2290 2291 2293 2295 [547] 2297 2298 2301 2302 2304 2305 2306 2307 2310 2311 2312 2314 2319 2320 2321 2324 2328 2329 2330 2332 2333 [568] 2335 2337 2339 2343 2344 2345 2346 2347 2348 2350 2353 2357 2358 2363 2365 2366 2367 2369 2370 2374 2375 [589] 2376 2378 2380 2384 2385 2387 2388 2390 2392 2393 2394 2399 2401 2403 2404 2405 2408 2409 2411 2412 2419 [610] 2422 2424 2429 2431 2432 2433 2438 2439 2447 2458 2461 2464 2477 2481 2486 2497 2498 2501 2594 2596 2597 [631] 2598 2623 2632 2636 2637 2648 2651 2653 2654 2656 2658 2661 2663 2664 2665 2666 2668 2672 2673 2674 2675 [652] 2676 2679 2680 2681 2683 2686 2687 2688 2689 2691 2692 2693 2695 2696 2697 2702 2704 2705 2706 2707 2708 [673] 2711 2712 2715 2716 2717 2718 2722 2723 2725 2726 2727 2728 2729 2730 2735 2736 2737 2738 2739 2740 2742 [694] 2743 2744 2745 2746 2747 2748 2749 2751 2753 2754 2755 2756 2758 2760 2761 2762 2763 2764 2765 2766 2767 [715] 2768 2771 2774 2775 2777 2778 2779 2780 2781 2782 2783 2784 2785 2787 2788 2789 2790 2791 2792 2796 2797 [736] 2798 2799 2801 2802 2803 2805 2808 2809 2810 2811 2812 2814 2815 2816 2818 2819 2820 2821 2823 2824 2825 [757] 2826 2827 2830 2832 2833 2836 2837 2838 2839 2841 2842 2843 2844 2845 2846 2847 2849 2851 2852 2853 2855 [778] 2856 2859 2861 2864 2865 2866 2867 2868 2870 2871 2872 2873 2874 2877 2878 2882 2884 2885 2887 2891 2892 [799] 2893 2895 2897 2900 2902 2903 2904 2905 2906 2908 2909 2910 2911 2912 2914 2915 2917 2922 2925 2926 2930 [820] 2931 2932 2934 2939 2940 2941 2942 2943 2944 2946 2947 2949 2955 2957 2958 2959 2960 2964 2965 2967 2968 [841] 2970 2971 2974 2975 2976 2977 2978 2979 2980 2982 2983 2985 2986 2987 2988 2989 2990 2991 2995 2996 2997 [862] 2999 3000 3001 3002 3003 3004 3006 3010 3013 3014 3015 3017 3018 3020 3021 3022 3027 3031 3032 3033 3034 [883] 3035 3036 3037 3039 3040 3041 3042 3043 3044 3045 3046 3047 3048 3049 3050 3051 3052 3053 3055 3056 3057 [904] 3058 3060 3061 3063 3067 3068 3070 3071 3072 3073 3075 3078 3080 3082 3083 3084 3085 3087 3089 3091 3097 [925] 3098 3100 3103 3104 3105 3110 3111 3113 3116 3120 3121 3133 3134 3135 3136 3137 3142 3143 3144 3145 3148 [946] 3149 3152 3154 3155 3156 3157 3158 3160 3161 3162 3163 3164 3165 3166 3170 3171 3172 3174 3176 3177 3183 [967] 3185 3193 3212 3213 3214 3226 3230 3232 3234 3235 3242 3248 3251 3252 3253 3256 3257 3260 3261 3263 3265 [988] 3266 3267 3268 3269 3271 3272 3275 3276 3278 3280 3281 3282 3287 [ reached getOption("max.print") -- omitted 631 entries ]

$purity min.abs.corr mean.abs.corr median.abs.corr L1 0.9851682 0.9995 1

$cs_index [1] 1

$coverage [1] 0.9501571

$requested_coverage [1] 0.95

[pcarbo]

Most likely this is an example where the association signal is not strong enough.

[PhoebeGuo97]

Hello @pcarbo, quite the opposite. All the SNPs's p values are ~1E-20 to ~1E-21.

[pcarbo]

The median abs. cor. is 0.9995 which is very high for this many SNPs. Maybe this is a very dense SNP panel or a SNP with unusually high LD?

[PhoebeGuo97]

I checked some literature and I think it's a region of high LD. In this case, I'm wondering how to prioritize these SNPs or if I should report all of them in the credible set. Many thanks!

[pcarbo]

In this case it is difficult to prioritize these SNPs; I would report the full CS (and general it is a good idea to report the full CS).

[PhoebeGuo97]

Thanks @pcarbo !