Advances in computational biology have occurred primarily in the areas of software and algorithm development; new designs of hardware to support biological computing are extremely scarce. This is due, we believe, to the presence of a non-trivial knowledge gap between molecular biologists and computer designers. The existence of this gap is unfortunate, as it has long been known that for certain problems, special-purpose computers can achieve significant cost/performance gains over general purpose machines. We describe one such computer here: a custom accelerator for gene sequence analysis. The accelerator implements a version of the Needleman-Wunsch algorithm for nucleotide sequence alignment. Sequence lengths are constrained only by available memory; the product of sequence lengths in the current implementation can be up to 2 22 . The machine is implemented as two NuBus boards connected to a Mac II f/x, using a mixture of TTL and FPGA technology clocked at 10 MHz. The boards are compl...

EFFICIENT ALGORITHMS FOR SEQUENCE ANALYSIS WITH CONCAVE AND CONVEX GAP COSTS David A. Eppstein We describe algorithms for two problems in sequence analysis: sequence alignment with gaps (multiple consecutive insertions and deletions treated as a unit) and RNA secondary structure with single loops only. We make the assumption that the gap cost or loop cost is a convex or concave function of the length of the gap or loop, and show how this assumption may be used to develop e#cient algorithms for these problems. We show how the restriction to convex or concave functions may be relaxed, and give algorithms for solving the problems when the cost functions are neither convex nor concave, but can be split into a small number of convex or concave functions. Finally we point out some sparsity in the structure of our sequence analysis problems, and describe how we may take advantage of that sparsity to further speed up our algorithms. CONTENTS 1. Introduction ............................1 ...

: We consider new algorithms for the solution of many dynamic programming recurrences for sequence comparison and for RNA secondary structure prediction. The techniques upon which the algorithms are based e#ectively exploit the physical constraints of the problem to derive more e#cient methods for sequence analysis. 1. INTRODUCTION In this paper we consider algorithms for two problems in sequence analysis. The first problem is sequence alignment, and the second is the prediction of RNA structure. Although the two problems seem quite di#erent from each other, their solutions share a common structure, which can be expressed as a system of dynamic programming recurrence equations. These equations also can be applied to other problems, including text formatting and data storage optimization. We use a number of well motivated assumptions about the problems in order to provide e#cient algorithms. The primary assumption is that of concavity or convexity. The recurrence relations for bo...