US Biomedical Research Communities

So far, my research career has taken me to UCLA, Harvard, and the University of Washington. For the next step of my research career, I could end up anywhere in the US. With this came the realization that I don’t actually know how representative these research environment / community experiences are to the rest of the potential places in the US I may end up working. Should I expect something entirely different in terms of resources, community, or culture?

To begin to address this, I looked to the data in the NIH RePORTER for statistics on the annual funding amounts awarded to each institution. I grouped the data by city, but was still not content as I knew this was not grouping the entirety of each research community (eg. Boston and Cambridge should be considered part of the same community). Thus, I called latitude/longitude coordinates for each city using the Data Science Toolkit, got all pairwise distances between all 819 cities, and clustered cities (by distance) using hierarchical clustering. I cut the tree to only include groups where the maximum distances between intra-group members were 50-miles, leaving me 229 separate clusters. I consider these to be separate “Research Communities”.

Here’s the corresponding map of the US where the major city from each cluster is shown as points, with each bubble sized according to its total funding amount, and the opacity of each bubble dependent on the total number of awards (which was highly correlated with total funding amount; see below).

Click on the image to see the map in better detail.

As you can tell by the map, there is a definite coastal bias, with some of the expected big players pretty obvious (eg. Boston area, San Francisco). I actually underestimated how much NIH money was in the New York City area, though it makes a lot of sense in retrospect. Here’s a table of the results:

groupcitystateannual funding (millions)awards
1BOSTONMA2647.3794044794
2NEW YORKNY1856.0498373648
3SAN FRANCISCOCA1613.6878893239
4DURHAMNC1065.6358591842
5LOS ANGELESCA1049.9809652009
6PHILADELPHIAPA1005.2615672088
7SEATTLEWA955.0843141522
8LA JOLLACA918.8439111710
9BALTIMOREMD897.3216281848
10CHICAGOIL717.1919491608
11PITTSBURGHPA625.3340951250
12HOUSTONTX564.1133591271
13ANN ARBORMI562.5307151249
14SAINT LOUISMO518.7724691072
15NEW HAVENCT466.434658994
16ATLANTAGA447.658441986
17NASHVILLETN388.092334849
18CLEVELANDOH332.186487675
19MADISONWI332.030422653
20MINNEAPOLISMN318.459912719
21WASHINGTONDC303.358981579
22BIRMINGHAMAL293.413142587
23PORTLANDOR274.705734534
24AURORACO253.082163641
25DALLASTX250.435769583
26BRONXNY249.238666498
27DAVISCA240.968518494
28COLUMBUSOH226.226589549
29ROCHESTERMN226.088176404
30SALT LAKE CITYUT209.105125513
31WORCESTERMA205.521402477
32CINCINNATIOH191.573358440
33CORAL GABLESFL187.671298397
34GAINESVILLEFL181.335999441
35PROVIDENCERI177.566364410
36IRVINECA177.126587362
37ROCHESTERNY169.69009398
38INDIANAPOLISIN169.278078355
39IOWA CITYIA168.922326403
40PISCATAWAYNJ162.599992382
41CHARLOTTESVILLEVA150.052283394
42BOULDERCO145.718439364
43LEXINGTONKY134.94632310
44CHARLESTONSC132.368946322
45WINSTON-SALEMNC130.540895255
46TUCSONAZ130.011104276
47MILWAUKEEWI128.291354282
48OMAHANE123.419242266
49SAN ANTONIOTX119.209378246
50KANSAS CITYKS114.48238265
51MEMPHISTN112.178482231
52AMHERSTNY109.40242264
53ROCKVILLEMD106.974511190
54NEW ORLEANSLA102.171044205
55HANOVERNH101.658468216
56DETROITMI100.722011232
57TEMPEAZ96.419453204
58ITHACANY95.211216245
59AUSTINTX89.629848298
60FARMINGTONCT87.73937225
61TAMPAFL85.787974175
62GALVESTONTX84.170354167
63RICHMONDVA83.614454238
64OKLAHOMA CITYOK82.806887177
65ALBUQUERQUENM82.01337163
66BAR HARBORME79.759078127
67CHAMPAIGNIL78.912638215
68UNIVERSITY PARKPA75.45165190
69EAST LANSINGMI73.48425193
70RESEARCH TRIANGLENC71.747136117
71STONY BROOKNY70.828814209
72LOUISVILLEKY68.366519165
73EVANSTONIL67.453466197
74COLLEGE STATIONTX65.806556203
75ATHENSGA65.735147155
76JUPITERFL61.823641122
77NEWARKDE60.758057122
78BURLINGTONVT56.939476120
79ALBANYNY55.686519142
80HERSHEYPA53.273163123
81EUGENEOR50.571811119
82WEST LAFAYETTEIN50.249558155
83LITTLE ROCKAR50.05940893
84COLUMBIAMO49.517631138
85AUGUSTAGA47.663342108
86PULLMANWA45.557797118
87COLUMBIASC43.46818398
88SANTA CRUZCA39.59948477
89TALLAHASSEEFL37.7234694
90BLOOMINGTONIN31.743822103
91RALEIGHNC31.669777102
92JACKSONVILLEFL30.68347849
93BLACKSBURGVA30.63630896
94RIVERSIDECA29.92964288
95JACKSONMS29.79006264
96BATON ROUGELA28.85699961
97ORANGEBURGNY28.38135664
98MORGANTOWNWV27.76834363
99SANTA BARBARACA26.22479868
100RENONV24.85050147
101SYRACUSENY24.2968978
102BOZEMANMT23.4506245
103NOVATOCA23.4338346
104EL PASOTX22.81792545
105NOTRE DAMEIN20.51245452
106PORTLANDME19.01048830
107CORVALLISOR18.06342848
108KINGSTONRI17.58929799
109GRAND RAPIDSMI16.34194932
110AMESIA16.34184451
111CLEMSONSC15.67978528
112SIOUX FALLSSD15.42730817
113ORLANDOFL15.40027843
114LUBBOCKTX15.09018344
115MARSHFIELDWI14.683875
116MANHATTANKS14.31095542
117MISSOULAMT14.29765837
118TOLEDOOH14.23106936
119HUNTSVILLEAL14.08251513
120LARAMIEWY13.85607116
121GRAND FORKSND13.46664720
122MONROVIACA12.56778723
123SHREVEPORTLA12.14681229
124MOBILEAL12.02287536
125RICHLANDWA11.70803517
126NORFOLKVA11.15285638
127WOODS HOLEMA10.938132
128LOMA LINDACA10.83018429
129LAS VEGASNV10.71934413
130LIVERMORECA10.31354224
131KNOXVILLETN9.43970427
132DAYTONOH9.26781729
133AUBURN UNIVERSITYAL9.0689724
134DANVILLEPA8.89004216
135FLAGSTAFFAZ8.09770710
136JOHNSON CITYTN7.7135723
137GREENVILLENC7.70407230
138HUNTINGTONWV7.6089477
139VERMILLIONSD7.58582818
140FARGOND7.56871315
141MERCEDCA7.02860522
142SPRINGFIELDIL6.95109511
143STILLWATEROK6.43771816
144BETHLEHEMPA6.41414218
145EDINBURGTX6.00239515
146BOISEID5.91432510
147TULSAOK5.4511896
148HATTIESBURGMS5.4264494
149UNIVERSITYMS5.31822213
150ST. PAULMN5.3094799
151MISSISSIPPI STATEMS4.78545810
152FAYETTEVILLEAR4.43391513
153LOS ALAMOSNM4.36796311
154ATHENSOH4.30814711
155TYLERTX4.129113
156DURHAMNH3.9379529
157DOVERDE3.7172998
158WICHITAKS3.5179683
159CARBONDALEIL3.34197511
160KALAMAZOOMI3.2032977
161DENTONTX2.72978210
162LOGANUT2.5257258
163WACOTX2.2725817
164LAFAYETTELA2.2155323
165FUQUAY VARINANC2.1497236
166MOUNT PLEASANTMI2.129317
167HOUGHTONMI1.954255
168SPRINGFIELDMO1.7597952
169SAN LUIS OBISPOCA1.5463076
170MELBOURNEFL1.4725523
171COLORADO SPRINGSCO1.4371855
172NORTH WEBSTERIN1.3728334
173WHITERIVERAZ1.3447072
174BILLINGSMT1.3389722
175SPOKANEWA1.1574814
176RAPID CITYSD1.1486822
177BELLINGHAMWA1.1087365
178DE KALBIL1.0662153
179FRESNOCA0.9736367
180DAVENPORTIA0.965523
181KINGSVILLETX0.9650455
182LAC DU FLAMBEAUWI0.9445311
183MACONGA0.9132882
184PEMBROKENC0.888023
185POTSDAMNY0.8438754
186BOWLING GREENKY0.8307982
187WILMINGTONNC0.8299242
188MIDDLEBURYVT0.8086392
189UTICANY0.7973331
190EL DORADO HILLSCA0.7700353
191SOCORRONM0.7683672
192TSAILEAZ0.7175711
193MONROELA0.7052262
194SPRINGFIELDVA0.6376921
195ALLENTOWNPA0.592582
196TEMECULACA0.563952
197PABLOMT0.5466092
198FORT SMITHAR0.5458452
199BURGESSVA0.5149132
200ODESSATX0.4977171
201FORT MYERSFL0.4930911
202KEENENH0.4161791
203SILVERTHORNECO0.40943
204OSHKOSHWI0.4075191
205GREAT FALLSMT0.4051
206KIRKSVILLEMO0.3823431
207DURANGOCO0.3599922
208BAKERSFIELDCA0.3566891
209SAVANNAHGA0.3532663
210STORM LAKEIA0.3296331
211KYKOTSMOVIAZ0.3078151
212TERRE HAUTEIN0.3053081
213SMYRNATN0.2992441
214ROLLAMO0.2800723
215DOTHANAL0.2772771
216BUENA VISTACO0.25611
217BUTTEMT0.2493071
218CHATTANOOGATN0.2367912
219WINTHROPME0.2321822
220MAGNOLIAAR0.2251
221SOUTH SAN FRANCISCOMI0.2249441
222BUFFALOWY0.2238481
223PENSACOLAFL0.1635763
224BEAUMONTTX0.13261
225ELIZABETH CITYNC0.1187811
226EAGLE BUTTESD0.0970191
227DAYTONA BEACHFL0.094751
228PITTSBURGKS0.0691141
229SCRANTONPA0.0585941

Surprisingly, Seattle came in at #7 in NIH funding, and Los Angeles came in at #5. With Boston the far-and-away #1, I guess this means I do indeed have a pretty biased view of research communities. That said, I’ve looked pretty deeply into a number of places already, and I suspect anything in the top 75 or so still have communities large enough to do what I want / need to do. And with things like video-conferencing now, it’s pretty easy to converse with people further away, anyways.

CAVEAT CAVEAT CAVEAT (My usual caveat statement):
1) Yes, I realize this is only NIH-funded award. Certainly other forms of national money like NSF or private money like HHMI matter a lot (and certainly from the individual investigator perspective), and are not taken into account my this analysis. Still, NIH-awarded money drives the brunt of biomedical research, so I don’t think this exclusion will lead to any overly misleading results.
2) Obviously, don’t take these rankings to mean anything about research community quality or anything. Just a data-driven metric to semi-quantitatively assess sheer research community size.

PS. Here’s the plot of award # vs total award amounts:

As you can see, it correlates pretty nicely, with a few outliers. Silverthorne is apparently where the Keystone Symposia are based. Marshfield is apparently the base of a regional health-care system that must have a handful of decently-sized grants.

EDIT 1:
By request, here’s a plot of the top 30 places (awarded greater than $10M annually) with the highest amount of money awarded per grant. I ended up making the 10M subset since there were a lot of random, small places which completely overtook the plot otherwise. Sometime in the future, after I figure out where to get population data, I’ll perform the other request where I divide area NIH awarded amount by area population.

EDIT 2:
OK, I had a number of different people express interest in having the funding amounts divided by city population, so I pulled data from the US census and did just that. A huge caveat is that this is ONLY looking at population of the “major” city in the group / cluster, and does not take into account the entire population of the cluster (that would require a fair bit more work, which I’m unwilling to devote right now). The results:

So there’s a pretty big cloud of points. I’ve highlighted the places with the most NIH funding per individual in the city (general population). You start seeing many more “college towns” pop up. Makes sense. Now all of the people at Yale and U Michigan can feel more content now. 😉

EDIT 3:
Thought I was done, but figured I’d do one more analysis I’m assuming some people care about. I subsetted on the most populous cities by the US census, and then looked at those cities NIH funding divided by the population. San Francisco doesn’t end up in this plot b/c of it’s small population (within city limits), so I let San Jose represent the bay area. Big caveat here is that I’m only using census data that represents population within the city, rather than metro areas.